Frequently asked questions
Every answer from across the site in one place - 670 questions covering verification, the free tools, and how RankCert compares to other launch platforms.
About RankCert
- What is RankCert?
- RankCert is a weekly launch board where products rank on one signal we verify ourselves - domain control - instead of upvotes. Every ranking links to dated evidence, and every signal is re-checked weekly.
- How is this different from an upvote board?
- On an upvote board the product with the largest existing network wins the day. Here, position is a function of what you can prove. Votes only break ties between products with the same proof.
- Does it cost anything to list?
- No. Listing and verification are free, and the link is dofollow whether or not you display our badge. We charge only for sponsor placement.
- What do you actually check?
- Unverified: Nothing checked yet. Verified: RankCert certificate live on the product's own domain.
- Do I have to connect any accounts?
- No. Verification is the RankCert certificate on your own domain. There is no analytics or payment connection to grant, and nothing to revoke.
- What happens if the proof disappears?
- Verification is re-checked. A certificate that is removed drops the product back to unverified at the next check.
Meta Tags Checker
- How long should a title tag be?
- Google truncates around 60 characters on desktop and roughly 50 on mobile. Under 60 keeps the whole title visible; under 20 usually means you are wasting the strongest ranking signal on the page.
- How long should a meta description be?
- Aim for 120 to 160 characters. Google rewrites descriptions about 70% of the time anyway, but a clear one still wins clicks when it is used.
- Does a missing canonical hurt rankings?
- It can. Without a canonical, the same content reachable at /page, /page?utm_source=x and /page/ competes with itself and splits its own link signals.
- Do you store the pages you check?
- We cache the result for an hour so repeat checks of the same URL do not re-fetch the site. We never store the page body.
- Which meta tags still matter in 2026?
- Title, meta description, canonical, meta robots and viewport. Everything else is either ignored (keywords, author, revisit-after) or handled better elsewhere. Open Graph tags matter for sharing, not for ranking.
- Does the meta keywords tag do anything?
- No. Google stopped using it in 2009 and Bing treats it as a spam signal. If a page still has one, delete it.
- Why does Google show a different title than mine?
- Google rewrites titles roughly a third of the time, usually when the title is truncated, keyword-stuffed, or does not match the query. A concise title under 60 characters that matches the H1 is rewritten least often.
- Should the title and H1 be identical?
- They should say the same thing but need not match word for word. The title competes for the click in the results page; the H1 confirms the topic once the visitor lands. Keep both about the same subject or you send a mixed signal.
Open Graph Preview
- What size should an og:image be?
- 1200 x 630 pixels, under 8MB, served over https. That ratio renders as a large card everywhere; anything smaller than 600 x 315 falls back to a small thumbnail.
- Why does my link still show the old image?
- Every platform caches Open Graph data. Use the platform's own debugger (Facebook Sharing Debugger, LinkedIn Post Inspector) to force a re-scrape after you change the tags.
- Do I need Twitter tags if I already have Open Graph?
- X falls back to og: tags, so you mostly need twitter:card set to summary_large_image. The other twitter: tags only matter if you want different copy on X.
- Does an og:image affect SEO?
- Not directly. It affects click-through on shared links, which is where its value is.
- Which Open Graph tags are required?
- og:title, og:type, og:image and og:url are the four in the specification. In practice og:description matters too, because most platforms show it under the title.
- Why does Slack show a card but LinkedIn does not?
- Each platform has its own crawler and its own cache. LinkedIn requires the image to be reachable without a redirect and at least 1200 x 627; Slack is far more forgiving. Run the LinkedIn Post Inspector to see its exact complaint.
- Can I use a different image per platform?
- Yes, via twitter:image for X and og:image for everything else, but it is rarely worth maintaining two. One 1200 x 630 image serves every platform well.
- Does the image need absolute URLs?
- Yes. Relative paths in og:image fail on most crawlers. Always use a full https:// URL.
WHOIS Lookup
- What is RDAP and why not WHOIS?
- RDAP is the IETF replacement for WHOIS. It returns structured JSON, supports internationalised data and is what registries actually maintain today. WHOIS text output is increasingly unreliable.
- Why is the registrant hidden?
- GDPR and registrar privacy services redact personal contact data for most domains. Registrar, dates, status and nameservers stay public.
- What does clientTransferProhibited mean?
- It is a registrar lock preventing the domain being transferred away without unlocking it first. It is a normal, healthy status, not a problem.
- Which TLDs are supported?
- Every TLD with an RDAP endpoint, which is all gTLDs and most ccTLDs. A handful of older ccTLDs are WHOIS-only and will return no record.
- Is a WHOIS lookup anonymous?
- The domain owner does not see who looked them up. The registry sees the query, as it would any DNS or RDAP request.
- How do I find out who owns a domain now that WHOIS is redacted?
- Check the registrar's own contact form, the domain's website for an imprint or privacy policy, and historical WHOIS archives from before May 2018. Registries will not release the data directly.
- What is the difference between the created and updated dates?
- Created is when the domain was first registered and never changes while the registration is continuous. Updated changes whenever any record on the domain changes - a nameserver edit, a renewal or a transfer.
- Can I tell if a domain is about to expire?
- Yes, the expiry date is public. Note that most registrars allow a redemption period of up to 75 days after expiry, so a domain is not actually available on the day it lapses.
Domain Age Checker
- Is domain age a Google ranking factor?
- Not on its own. Google has said age itself carries little weight. What correlates with age is accumulated links and history, which do matter - an old domain with no links ranks no better than a new one.
- Does buying an aged domain help?
- Only if its link profile is real and topically relevant. A dropped domain loses most of its value once it changes ownership and topic.
- Why does the age differ from other tools?
- Many tools report first-seen-in-archive rather than registration date. We report the registry's own registration event, which is the authoritative number.
- What if the registration date is missing?
- Some ccTLD registries do not publish registration events over RDAP. In that case we show the dates the registry does return.
- How old is old enough for a domain to rank?
- There is no threshold. New domains rank for low-competition queries within weeks. What takes time is accumulating the links and history that competitive queries need, and that is a function of work, not calendar age.
- Does letting a domain expire reset its age?
- If it drops and is re-registered, the registry records a new creation date and the domain is treated as new. Renewing before it drops changes nothing.
- Is there a Google sandbox for new domains?
- Google has denied a formal sandbox. New sites do rank more slowly, which is more plausibly explained by having no links, no click history and little crawl priority than by a deliberate penalty.
- Where can I check what a domain used to be?
- The Internet Archive's Wayback Machine. Before buying an aged domain, look at what it hosted - a former spam or adult site carries baggage no amount of new content clears quickly.
UTM Builder
- Which UTM parameters are required?
- utm_source, utm_medium and utm_campaign. GA4 will record a visit with only some of them set, but attribution breaks down without all three.
- Should UTM values be lowercase?
- Yes. UTM values are case-sensitive, so Twitter and twitter become two rows in your report. Pick lowercase and never deviate.
- What is the difference between source and medium?
- Source is where the click came from (newsletter, producthunt, x). Medium is the type of channel (email, referral, social, cpc). Source answers who, medium answers how.
- Do UTM parameters hurt SEO?
- Only if a tagged URL gets indexed. Set a canonical to the clean URL and the duplicate never competes.
- Do I need UTM tags on internal links?
- No, and it actively breaks reporting. An internal link with a utm_source starts a new session and credits the visit to itself, wiping out the original acquisition source.
- What are utm_term and utm_content for?
- utm_term carries the paid keyword, utm_content distinguishes two links in the same email or ad that point at the same place. Both are optional and only useful if you will actually read them.
- How long do UTM values persist?
- GA4 attributes the session at entry. The parameters do not follow the visitor to later pages, which is why they belong only on the first click.
- Can I shorten a UTM link?
- Yes - a shortener passes the parameters through on redirect. Just make sure the shortener uses a 301 or 302 and does not strip the query string.
JWT Decoder
- Does my token leave the browser?
- No. Decoding happens in JavaScript on this page. There is no network request and no server involved, which is why this tool works offline once loaded.
- Does this verify the signature?
- No. Verifying requires the signing secret or public key, and you should never paste a secret into a website. Decoding shows you the claims; verification belongs in your own code.
- What do exp, iat and nbf mean?
- exp is expiry, iat is issued-at, nbf is not-valid-before. All three are NumericDate values - seconds since the Unix epoch, not milliseconds.
- Is a JWT encrypted?
- No. A standard JWS token is signed, not encrypted, and anyone holding it can read the payload. Never put secrets in a JWT.
- Can anyone read my JWT?
- Yes. A standard JWT is signed, not encrypted, so anyone who intercepts it reads every claim. Treat it as a bearer credential: transmit over https, store carefully, and put no secrets in the payload.
- What is the difference between JWS and JWE?
- JWS is signed and readable - it is what almost everyone means by JWT. JWE is encrypted and has five segments instead of three. If your token has four dots, it is JWE and cannot be decoded without the key.
- Why does my token say invalid when it looks fine?
- Usually padding or whitespace. A JWT copied from a log or header often picks up a trailing newline or a Bearer prefix. Strip both before decoding.
- How do I check whether a token has expired?
- Read the exp claim, which is seconds since the Unix epoch, and compare it to now. A token expiring in 1,700,000,000 is not the year 1970 - it is a seconds value that needs multiplying by 1000 for JavaScript Date.
JSON Formatter
- Is my JSON uploaded anywhere?
- No. Parsing and formatting run in your browser with the native JSON API. Nothing is sent to a server.
- Why does my JSON fail to parse?
- Almost always one of three things: a trailing comma, single quotes instead of double quotes, or an unescaped newline inside a string. JSON allows none of them.
- What indent should I use?
- Two spaces is the common default for config and API payloads. Use minified output for anything you transmit - it is usually 15 to 30% smaller.
- Does formatting change my data?
- No, but key order is preserved and duplicate keys collapse to the last value, which is standard JSON.parse behaviour.
- What is the difference between JSON and JSON5?
- JSON5 allows comments, trailing commas, single quotes and unquoted keys. It is convenient for config files and rejected by every standard JSON parser, including this one.
- Can JSON have comments?
- No. The specification has no comment syntax. Config formats that appear to allow them (tsconfig, VS Code settings) use JSONC, a superset their own parsers handle.
- Why do my large numbers change when formatted?
- JavaScript parses all numbers as 64-bit floats, so integers above 2^53 lose precision. If your API returns large IDs, have it send them as strings.
- How do I format JSON from the command line?
- Pipe it through jq: curl -s url | jq . formats and colourises. Node also has it built in: node -e "console.log(JSON.stringify(require('./f.json'),null,2))".
UUID Generator
- What is the difference between UUID v4 and v7?
- v4 is 122 random bits. v7 puts a millisecond timestamp in the leading 48 bits, so IDs sort chronologically. Both are 128 bits and collision-safe in practice.
- Should I use a UUID as a primary key?
- Use v7, not v4. Random v4 keys scatter inserts across a B-tree index, causing page splits and bloat. v7 appends in order, so it behaves much more like a sequential key.
- What are the odds of a collision?
- For v4, you would need roughly 2.7 x 10^18 UUIDs before a 50% chance of one collision. It is not a practical concern.
- Are these random enough for security tokens?
- They come from crypto.getRandomValues, so yes for identifiers. For session or reset tokens, generate them server-side and store a hash.
- Is UUID v7 safe to use in production yet?
- Yes. It was standardised in RFC 9562 in 2024 and is supported by Postgres extensions, most ORMs and every major language's UUID library.
- Does a UUID v7 leak information?
- It leaks the creation time to the millisecond, since the timestamp is in plain sight. That is usually harmless, but do not use v7 where the creation time is sensitive.
- UUID or ULID?
- They solve the same problem. ULID is 26 characters in Crockford base32 and sorts lexically as a string; UUID v7 is the standardised, more widely supported equivalent. Prefer v7 unless you need the shorter text form.
- Should UUIDs be stored as text?
- No. Store them as a native uuid type in Postgres or a 16-byte binary column in MySQL. Storing as a 36-character string roughly doubles the index size for no benefit.
SEO Checker
- What does the score mean?
- It is the share of eleven on-page checks the page passes. It is a completeness measure, not a ranking prediction - a page can score 100 and still not rank if it has no links or no depth.
- Does this replace a full site crawl?
- No. This audits one URL. Site-wide problems - orphan pages, duplicate titles across templates, broken internal links - need a crawler.
- Why does word count matter?
- Under about 300 words of visible text, a page rarely has enough substance to rank for anything competitive. It is a rough signal, not a target to pad toward.
- Do you run JavaScript?
- No. We read the HTML your server returns, which is what a crawler reads on its first pass. If your content only appears after JavaScript runs, that is worth knowing.
- How often should I re-run an on-page check?
- After any template change, and once a quarter otherwise. Most on-page regressions arrive with a deploy, not with time.
- What is the single highest-impact on-page fix?
- A title that matches the query the page targets. Nothing else on the page moves rankings as reliably, and it is a one-line change.
- Does this check Core Web Vitals?
- No. Field performance data comes from Chrome's CrUX report and needs real user traffic. Use PageSpeed Insights for that side.
- Why does a page pass every check and still not rank?
- On-page work makes a page eligible, not competitive. If the pages ranking above you have hundreds of referring domains and yours has none, the gap is authority, not tags.
Redirect Checker
- 301 or 302?
- 301 and 308 are permanent and pass link signals. 302 and 307 are temporary and tell search engines to keep indexing the original URL. Using 302 for a permanent move is one of the most common and costly redirect mistakes.
- How many redirects are too many?
- One. Every extra hop adds latency and costs crawl budget, and Google stops following after about five. A chain of three should be flattened to one.
- Do redirects lose link equity?
- Modern Google passes essentially full PageRank through 301s, so the loss is not the issue. Crawl budget, latency and the risk of a broken middle hop are.
- What is a redirect loop?
- A URL that eventually redirects back to itself. Browsers give up after about twenty hops and show an error; crawlers give up much sooner and drop the URL.
- Does a redirect chain hurt rankings?
- Indirectly. Google follows up to about five hops, so a long chain risks the destination never being reached, and every hop adds real latency for users. Flatten chains to one hop.
- What is the difference between a 307 and a 302?
- Both are temporary. A 307 guarantees the request method is preserved, so a POST stays a POST. A 302 historically allowed clients to convert POST to GET, which some still do.
- Should I redirect http to https at the server or the CDN?
- At the edge, before the request reaches your origin. It is faster and it means the redirect still works if the origin is down.
- How long should I keep old redirects?
- At least a year, and indefinitely if the old URL still has external links. Redirects are cheap; broken links are not.
Broken Link Checker
- Do broken links hurt SEO?
- Broken internal links do. Every one is link equity poured into a dead end and crawl budget spent on nothing. Broken external links do not carry a direct penalty, but a page full of rotted citations reads as unmaintained to both readers and quality raters.
- How many links does this check?
- The first 50 links on the page. That covers almost every article and most navigation-heavy pages, and it keeps the scan under the timeout instead of hanging for a minute.
- Does it crawl my whole site?
- No - it crawls one page at a time. Run it on your sitemap's busiest pages, or on your blog index to sweep every post link at once.
- Why does a link show as dead when it opens fine in my browser?
- Some sites block unknown user agents or answer HEAD requests with an error. We retry with a normal GET when a server rejects HEAD, but a hard bot block still reads as dead. Open anything surprising by hand before deleting it.
- What is the difference between a broken link and a redirect?
- A redirect still resolves - it just costs a hop. A broken link answers 4xx or 5xx and resolves to nothing. Use the redirect checker for chains and this for dead ends.
- Should I fix or delete a dead external link?
- Look it up in the Wayback Machine first. If the page still exists somewhere, repoint it. If the source is gone for good, remove the link and keep the sentence.
- How often should I run a broken link audit?
- Quarterly for a small site, monthly if you publish weekly. External link rot runs at roughly 5% a year, so an old post quietly accumulates dead citations without you touching it.
- Why are http links flagged?
- They are not broken, but they force a downgrade and an extra redirect hop on every click. Updating them to https is a one-line fix that removes a hop from the chain.
Backlink Verifier
- How do I check if a backlink is live?
- Fetch the page the link is supposed to be on and look for an anchor pointing at your domain in the raw HTML. That is exactly what this tool does, and it is what a crawler does too - opening the page in your browser is a weaker test, because your browser runs JavaScript that Googlebot may never execute.
- What is the difference between dofollow and nofollow?
- A dofollow link carries no rel restriction, so it passes ranking signal to the target. A rel="nofollow", rel="sponsored" or rel="ugc" link is a hint to search engines not to pass that signal. Both kinds send real visitors and both count as brand mentions.
- My link is there but shows as nofollow. Is it worthless?
- No, but it is worth less. It still sends referral traffic, still gets your brand in front of readers, and still turns up in link graphs that AI systems read. It just will not move rankings the way a dofollow link does.
- Why does the tool say no link when I can see it on the page?
- Two usual causes. The page renders its links with JavaScript, so they do not exist in the HTML a crawler receives - which means search engines may miss the link too. Or the link points at a redirector or a different domain spelling than the one you entered.
- What does noindex on the linking page mean for my link?
- It means the page is not in Google's index, so there is no page from which signal can flow. Directories sometimes noindex their listing pages while still advertising dofollow links. Ask for placement on an indexable page instead.
- Can a page-level nofollow cancel my dofollow link?
- Yes. A meta robots nofollow, or an X-Robots-Tag nofollow header, applies to every link on the page regardless of the rel on each anchor. This tool flags it when it sees one.
- How long after a link goes up should I check?
- Give it a few minutes for their CDN cache to clear, then check. Re-check a month later - links quietly disappear during redesigns, and paid or swapped placements get removed more often than anyone admits.
- Does the anchor text matter?
- It is one of the stronger relevance signals a link carries. A descriptive anchor beats "click here", and an image link with no alt text carries no keyword signal at all. This tool shows the anchor on every match so you can see what you actually got.
Outbound Link Checker
- What is an outbound link?
- Any link on your page that points at a different domain. Links to other pages on your own site are internal links, which move equity around inside your site rather than out of it.
- Do outbound links hurt SEO?
- Relevant outbound citations help - they are a normal part of a well-sourced page and Google has said as much. What hurts is handing dofollow links to sites you would not vouch for, or selling links without marking them sponsored.
- When should I use rel="sponsored"?
- On any link you were paid for, including affiliate links. Google's guidance is explicit: paid links that pass equity are a link scheme, and rel="sponsored" is the way to keep the link while declaring the arrangement.
- What is rel="ugc" for?
- Links inside user-generated content - comments, forum posts, profile fields. It tells search engines a visitor wrote this link, not you, which limits the damage when spammers find your comment form.
- How many outbound links is too many?
- There is no hard cap, but a page handing out dozens of dofollow links to unrelated domains reads as a link farm. Keep the citations that support what you wrote and nofollow or drop the rest.
- Why check outbound links on someone else's page?
- Before you buy or swap a placement. A page that already passes dofollow equity to fifty unrelated sites will pass you very little, and a page carrying a site-wide robots nofollow will pass you nothing at all.
- Does this find links added by JavaScript?
- No, and that is deliberate. It reads the same HTML a crawler receives, so what you see here is what search engines see. Links that only appear after JavaScript runs are links you should not count on.
- Does an http:// outbound link matter?
- It is not broken, but it costs a redirect hop on every click and downgrades the connection before the target upgrades it again. Updating to https is a one-line fix.
DNS Lookup
- What is a TTL?
- Time to live, in seconds - how long resolvers may cache the record. Lower it to 300 a day before a migration so the change propagates quickly, then raise it back.
- Why do I see different records than my registrar shows?
- You are seeing what public resolvers currently have cached. Immediately after a change, the old value can persist for up to the previous TTL.
- A record or CNAME for the apex?
- The DNS spec does not allow a CNAME at the apex alongside other records. Use an A record, or your provider's ALIAS/ANAME flattening if you need to point the apex at a hostname.
- What are the TXT records for?
- Domain verification, SPF, DKIM and DMARC. If your email is landing in spam, the SPF and DMARC TXT records are the first place to look.
- How long does DNS propagation take?
- Up to the previous TTL of the record, which is why you lower it to 300 seconds a day before a change. There is no global propagation queue - each resolver simply holds its cached copy until it expires.
- What is an SOA record for?
- It names the primary nameserver and holds the zone's serial number and refresh timers. A serial that never changes is a sign the zone is not being updated where you think it is.
- Why do MX records have numbers in front?
- That is the priority. Lower wins. Mail servers try the lowest-numbered host first and fall back to higher ones, so two records at the same priority load-balance.
- How do I check DNS from a terminal?
- dig +short example.com A, or dig example.com MX for mail. Add @1.1.1.1 to query a specific resolver and bypass your local cache.
Domain Popularity Checker
- Is this Ahrefs DR or Moz DA?
- Neither, and it does not try to be. Those are backlink scores sold behind a subscription. This is a popularity score derived from Tranco, an open research ranking of the top million domains by traffic. It answers a different question - how many people actually visit this site - and it is free.
- What does the score mean?
- It maps the Tranco rank onto 0-100 on a log scale: rank 1 scores 100, rank 1,000 scores about 50, rank 1,000,000 scores 0. Most working SaaS sites land between 20 and 50.
- Why does my domain score zero?
- Because it is outside the top million domains by traffic. That is normal for a young product and says nothing about content quality - it just means the popularity data has nothing to report yet.
- Is popularity a Google ranking factor?
- No. Google does not use Tranco or any third-party score. It is useful as a rough sanity check on how big a competitor really is, not as a target to optimise.
- How is this different from Ahrefs DR?
- Ahrefs DR measures backlinks from a proprietary crawl. This measures traffic from an open research ranking. A site can score high on one and low on the other, and neither is a Google signal.
- What score should I aim for?
- None - it is not a target. Use it as a comparison: if every site ranking for your keyword scores above 55 and you score 10, the keyword is a long project rather than a quick win.
- How often does the underlying ranking update?
- Tranco publishes a new list daily, averaged over the preceding 30 days, which is what makes it stable enough to compare week to week.
- Does a high popularity score mean a link from that site is valuable?
- It means people visit, which makes the link worth having for referral traffic alone. Whether it passes ranking value depends on relevance and whether the link is followed.
Bulk Domain Popularity
- What should I do with the list?
- Triage it. A directory sitting outside the top million rarely returns the hour it takes to submit; anything scoring above 45 is worth a personal pitch.
- Why only 25?
- Each domain is a separate lookup against a public ranking service. Twenty-five keeps the request fast and keeps us a polite client of a free service.
- Do results get cached?
- Yes, for a week. Tranco itself only republishes daily, so a fresher lookup would return the same number.
- What is a good use for a bulk check?
- Directory submission lists and link prospect lists. Both are usually padded with dead sites, and one pass tells you which two dozen entries out of a hundred are worth an hour.
- Can I export the results?
- Copy them straight out of the table - the values are plain text and paste into a spreadsheet cleanly.
- Why do some domains return nothing?
- They sit outside the top million by traffic. That is the honest answer rather than a fabricated score, and for a directory it usually means skip it.
- Does subdomain or www matter?
- Enter the registrable domain without www. The ranking is measured at the domain level, so blog.example.com and example.com return the same figure.
Bulk Domain Checker
- How many domains can I check at once?
- Twenty-five per run. That is the point where registries start rate-limiting a single caller, and a slower tool that gets throttled is worse than a fast one that respects the limit. Run it again for the next batch.
- Is this the same as a WHOIS lookup?
- It uses RDAP, the protocol ICANN mandated to replace WHOIS. RDAP returns structured JSON instead of scraped text, so registrar and expiry come back reliably rather than parsed out of free-form output.
- Does it work for every TLD?
- For every TLD that publishes an RDAP service - all gTLDs and most ccTLDs. A few registries, .de among them, still do not. Those come back marked unconfirmed rather than guessed at, because a false 'available' is the one answer this tool must never give.
- Does checking a domain here risk front-running?
- No. RDAP queries go to the registry, not to a registrar's search box, and nothing you type is logged against your session or sold on. Front-running has historically come from registrar search tools, not registry lookups.
- Why does a domain show as available when a registrar says it is taken?
- Usually a premium or reserved name: unregistered at the registry, but withheld from general sale. RDAP reports the registration status, which is the truth about who holds it, not the sales status.
- What does the expiry date mean for a taken domain?
- It is when the current registration lapses, not when the name becomes buyable. Expect a 30-day grace period plus a 30-day redemption window after that, and most owners renew. Treat it as a signal, not a date to plan around.
- Can I check availability without the TLD?
- No - example.com and example.io are different registrations at different registries. List every extension you actually want, one per line.
- Is there a limit on how often I can run it?
- Four bulk runs a minute from one IP. Results are cached for an hour, so re-checking the same list costs nothing.
Keyword Idea Generator
- Where do the ideas come from?
- Google autocomplete, the same suggestions that drop down in the search box. They are built from real queries people type, so every idea here is a phrasing somebody has actually searched.
- Why is there no search volume?
- Because honest volume data costs money, and a made-up number is worse than none. Autocomplete tells you a query exists and how it is phrased. For demand, publish the page and read the impressions in Search Console.
- How should I use the angles?
- Question phrasings suit a guide or FAQ block, comparison and alternative phrasings suit a dedicated comparison page, and commercial phrasings belong on the page that sells the thing.
- How do I pick which ideas to write about?
- Take the question phrasings your product can answer better than a generic guide. Long, specific queries have less competition and convert far better than the two-word head term.
- Do I need search volume to start?
- No. For a new site, volume is the wrong filter - almost everything with volume is out of reach. Write for specific questions first, then check volume once you have rankings to defend.
- Are autocomplete suggestions personalised?
- Lightly, by country and language. The core suggestions are aggregate query data, which is why they are a reasonable proxy for demand.
- What is the difference between this and a keyword tool with volume?
- Paid tools model volume from clickstream data. This shows demand exists without pretending to know its size. For content planning, knowing the exact phrasing matters more than a number with a wide error bar.
robots.txt Generator
- Does robots.txt stop a page being indexed?
- No. It stops crawling, not indexing. A blocked URL with external links can still appear in results with no snippet. To keep a page out of the index, allow crawling and use a noindex meta tag.
- Should I block AI crawlers?
- Usually not. Blocking GPTBot or ClaudeBot removes you from AI answers as well as training data, and for a small product the citations are worth more than the exclusion.
- Does Crawl-delay work?
- Google ignores it. Bing and Yandex honour it. Only set it if your server is genuinely struggling.
- Where does the file go?
- Exactly at yourdomain.com/robots.txt, served as text/plain. Subdirectories do not work - each subdomain needs its own.
- What should a normal site block?
- Admin, checkout, account and internal search result pages. Nothing else. Blocking CSS and JavaScript stops Google rendering your page and is a common self-inflicted wound.
- Does robots.txt need to be at the root?
- Yes, exactly at yourdomain.com/robots.txt. A file at /blog/robots.txt is ignored entirely, and each subdomain needs its own.
- How do I keep a page out of Google properly?
- Use a noindex meta robots tag and let the page stay crawlable. Blocking it in robots.txt means Google cannot read the noindex and may index the URL anyway from external links.
- What happens if robots.txt returns a 500 error?
- Google treats a server error as a full disallow and may stop crawling the site entirely. A 404 is safe - it is read as allow everything.
llms.txt Generator
- Does llms.txt actually do anything yet?
- It is a proposal, not a standard, and no major AI engine has committed to reading it. It costs ten minutes and is a plain-language summary of your site, so the downside is negligible and the upside is real if adoption comes.
- How is it different from robots.txt?
- robots.txt says what crawlers may not fetch. llms.txt says what a model should read first and how to describe you. One restricts, the other guides.
- What should go in it?
- A one-line description, then links to the pages that answer the questions people ask about your product - pricing, docs, comparisons. Not your whole sitemap.
- Should I also serve markdown versions of pages?
- The full proposal suggests a .md twin of each page. That is more work and more useful - it gives the model clean text with no navigation chrome.
- Do AI engines actually read llms.txt?
- No major engine has confirmed it as a ranking or retrieval input. The pages it points at are read regardless, so the file is cheap insurance rather than a lever.
- What is llms-full.txt?
- An optional companion holding the full text of your key pages in one file, so a model can ingest everything without following links. Worth adding if your site is small.
- How does this compare with an XML sitemap?
- A sitemap is exhaustive and machine-formatted for crawl scheduling. llms.txt is curated and human-readable - a shortlist of what matters with a sentence explaining each entry.
- Should I include my blog posts?
- Only the ones that answer real questions about your product or field. A dump of every post defeats the purpose, which is curation.
Schema Markup Generator
- JSON-LD, microdata or RDFa?
- JSON-LD. Google explicitly recommends it, and it lives in one block in the head instead of being tangled through your markup.
- Does schema improve rankings?
- Not directly. It changes how your result is displayed and makes your content far easier for AI answer engines to parse and cite, both of which affect clicks.
- Can I use more than one type on a page?
- Yes. A blog post typically carries Article plus BreadcrumbList, and often FAQPage. Use separate script tags or an @graph array.
- How do I know it is valid?
- Valid JSON is not valid schema. Run it through Google's Rich Results Test, which checks required properties rather than just syntax.
- Which schema types are worth adding first?
- Organization on the homepage, Article or BlogPosting on posts, BreadcrumbList site-wide, and Product with offers if you sell. Those four cover nearly all the rich result opportunity most sites have.
- Can schema get me penalised?
- Yes, for marking up content that is not on the page or is misleading. Google issues manual actions for structured data spam. Mark up what is visible and you are safe.
- Does schema help with AI search?
- It helps models parse entities and relationships reliably rather than inferring them from prose. It is not a citation guarantee, but ambiguous pages get cited less.
- Where should the JSON-LD block go?
- In the head, or anywhere in the body - both are valid. Keeping it in the head next to the other metadata is easier to maintain.
FAQ Schema Generator
- Does FAQ schema still get rich results?
- Google restricted FAQ rich results to authoritative government and health sites in 2023. The markup is still worth adding because AI answer engines parse it directly when deciding what to quote.
- Can I mark up questions that are hidden?
- No. The content must be visible on the page. Marking up hidden or accordion-only-after-JS content risks a structured data manual action.
- How many questions should I include?
- Four to six, each answering a real question people ask. Padding with invented questions dilutes the page and helps nothing.
- Where should the script go?
- Anywhere in the document - head or body both work. Keeping it in the head next to your other JSON-LD is easier to maintain.
- Is FAQ schema still worth adding?
- For rich results, only if you are a government or health authority. For everything else it is worth adding because it gives crawlers and language models unambiguous question-answer pairs to lift, which matters more each year.
- Can two pages carry the same FAQ block?
- They can, but duplicating the same questions site-wide dilutes the value and looks templated. Write questions specific to each page.
- Does FAQPage conflict with other schema types?
- No. A page can carry Article, BreadcrumbList and FAQPage simultaneously in one @graph or as separate script blocks.
- How long should each answer be?
- One to three sentences that fully answer the question. An answer that ends in read more is not an answer and gets ignored.
Word Counter
- How is reading time calculated?
- At 225 words per minute, the average adult silent reading speed for non-technical prose. Speaking time uses 130 words per minute, the usual presentation pace.
- Does my text get stored?
- No. Everything is counted in JavaScript on this page. There is no network request, which is why it keeps working offline.
- How long should a blog post be?
- Long enough to answer the question completely. In practice, pages ranking for competitive terms tend to run 1,200 to 2,000 words - not because length ranks, but because complete answers are long.
- What counts as a word?
- Any run of non-whitespace characters. Hyphenated compounds count as one, which matches how most editors and word processors count.
- How many words is a typical page of writing?
- About 250 double-spaced or 500 single-spaced in 12pt. A 2,000-word article is roughly eight double-spaced pages.
- Does character count include spaces?
- Both figures are shown. Social platforms count characters with spaces - X allows 280, a meta description should stay under about 160.
- How many words should an essay or article be?
- Whatever fully answers the question. For search, the pages ranking for a query are the honest benchmark: match their depth rather than a word target.
- Can I count words offline?
- Yes - once this page has loaded it needs no network. The counting runs entirely in JavaScript on your device.
Reading Time Calculator
- What reading speed should I use?
- 225 words per minute for general prose. Technical content with code is closer to 150, so round up for developer documentation.
- Does showing reading time help?
- It sets expectations, which reduces bounces on long posts. It is not a ranking factor, but engagement effects are real.
- Should code blocks count?
- Not at prose speed. If a post is a third code, add roughly 50% to the estimate.
- How long should a post be?
- As long as the answer needs. A 4-minute post that fully answers the question beats a padded 12-minute one.
- What reading speed do most sites use?
- Between 200 and 265 words per minute. Medium famously uses 265, which flatters long posts. 225 is closer to measured adult silent reading speed for non-technical prose.
- Does reading time affect SEO?
- Not directly. It affects whether people start reading, which affects engagement, which affects everything downstream.
- Should I round up or down?
- Round up and never show zero. A 40-second read should say 1 min, not under a minute, which reads as filler.
- How do I account for images and code?
- Add about 10 seconds per image and count code at roughly half prose speed. A tutorial that is a third code needs about 50% more time than the word count suggests.
Keyword Density Checker
- Is there an ideal keyword density?
- No. Google has not used density as a signal for well over a decade. The tool is useful in one direction only: catching accidental over-repetition that reads badly to humans.
- What density is too high?
- Above about 3% for a single term usually reads as stuffing. The fix is to rewrite the sentences, not to swap in synonyms mechanically - that reads worse.
- Why are common words excluded?
- Articles, prepositions and pronouns dominate any text and tell you nothing. They are filtered so the list shows meaningful terms.
- Should my target keyword be the top word?
- Usually, but only just. If your target phrase is not in the top few at all, the page probably is not about what you think it is about.
- What is a natural keyword density?
- Whatever falls out of writing normally, usually between 0.5% and 2% for the main topic. If you have to check whether you used the phrase enough, you probably used it too much.
- Does keyword stuffing still get penalised?
- Yes. It is an explicit spam policy violation, and beyond the penalty it reads badly, which costs conversions long before it costs rankings.
- What should I look at instead of density?
- Whether the page covers the subtopics the ranking pages cover. Coverage of related concepts predicts rankings far better than repetition of one phrase.
- Why does the tool show two-word and three-word phrases?
- Because real queries are phrases. Seeing that best time tracking software appears nine times is more useful than seeing software appears forty.
Readability Checker
- What Flesch score should I aim for?
- 60 to 70 for general audiences - plain English, around 8th to 9th grade. Technical documentation naturally lands at 30 to 50 and that is fine for a developer audience.
- Is readability a ranking factor?
- Not directly. It affects how long people stay and whether they finish, and those behavioural signals do matter.
- How is the score calculated?
- From average sentence length and average syllables per word. Long sentences hurt the score roughly twice as much as long words do, which is why splitting sentences is the fastest fix.
- Should I write for the lowest score possible?
- No. Below about grade 6 the writing becomes choppy and condescending. Clarity is the goal, not simplicity.
- What Flesch score do popular sites hit?
- Most successful consumer writing lands between 60 and 80. Legal and academic writing sits below 30, which is why it is exhausting to read.
- Does simplifying language mean dumbing down?
- No. It means shorter sentences and fewer Latinate words, not fewer ideas. Technical accuracy and readability are independent - you can explain a complex thing in short sentences.
- Why did my score drop when I added a code block?
- Syllable counting treats identifiers and URLs as very long words. Score the prose, not the code.
- Which single change improves readability most?
- Splitting sentences over about 25 words. Sentence length dominates both formulas and is the easiest thing to fix mechanically.
Flesch Reading Ease Calculator
- What is a good Flesch Reading Ease score?
- 60 to 70 for general web writing, marketing copy and documentation. Below 50 you are writing for a university-educated reader whether you meant to or not. Above 80 can read as simplistic for a technical audience.
- How is the Flesch Reading Ease score calculated?
- 206.835 - 1.015 x (words per sentence) - 84.6 x (syllables per word). That is the whole formula. It measures sentence and word length, nothing else.
- What is the difference between Flesch Reading Ease and Flesch-Kincaid?
- Same two inputs, opposite direction. Reading Ease gives a 0-100 score where high is easy. Flesch-Kincaid gives a US school grade level where high is hard. Use Reading Ease for a target to hit, Flesch-Kincaid when someone asks for a grade number.
- Does a high Flesch score mean the writing is good?
- No. The formula cannot see whether a sentence is true, clear or worth reading - only how long it is. A page of disconnected short sentences scores 90 and reads badly. Treat it as a check on density, not a quality grade.
- Does readability affect SEO?
- Not as a ranking factor. It affects whether people finish the page, and Google has said plainly that there is no readability score in the ranking systems. Write for the reader; the score is a proxy for whether you did.
- Why did my score change when I only fixed punctuation?
- The formula counts sentences by terminal punctuation. Adding a full stop splits one long sentence into two shorter ones, which moves the score more than any word swap will.
- How does this tool count syllables?
- With the standard vowel-group heuristic that every browser-based implementation uses - it groups adjacent vowels and strips common silent endings. It is right the large majority of the time and occasionally off on unusual names, which is enough for a score you are reading as a band, not a decimal.
- How much text do I need for a reliable score?
- At least 100 words. Under about 30 words a single long sentence swings the result by twenty points or more, so short snippets give you noise.
Flesch-Kincaid Grade Level Calculator
- What grade level should I write at?
- Seventh to ninth grade for anything aimed at a general audience, including most B2B marketing - the reader's education level is not the constraint, their attention is. Regulated plain-language rules often demand sixth to eighth.
- How is Flesch-Kincaid Grade Level calculated?
- 0.39 x (words per sentence) + 11.8 x (syllables per word) - 15.59. Note the weighting: syllables per word carries roughly thirty times the coefficient, but sentence length varies far more in real writing, so in practice sentence length is the lever.
- Is Flesch-Kincaid the same as Flesch Reading Ease?
- Same inputs, different output. Reading Ease is 0-100 and higher is easier; Grade Level is a school year and higher is harder. They are recalibrations of one another, not independent measurements.
- Can the grade level exceed 12?
- Yes, and routinely does. Insurance policies and academic abstracts often land between 15 and 20. There is no ceiling in the formula.
- Why do US agencies use Flesch-Kincaid?
- The US Navy developed it in 1975 to grade training manuals, and it was written into federal plain-language standards from there. That institutional history, not superior accuracy, is why it is the one most often specified in a contract or style guide.
- Does it work for languages other than English?
- No. The coefficients were fitted to English syllable and sentence patterns. German or Finnish text will return a number, and that number is meaningless. Use a formula calibrated for the language.
- Why does my technical writing grade so high?
- Multi-syllable domain terms - implementation, authentication, infrastructure - drive syllables per word up, and the formula cannot tell a jargon word your reader knows cold from one they have never seen. For a specialist audience, a high grade may be correct rather than a problem.
- How much text should I paste in?
- At least 100 words. Below roughly 30, one long sentence moves the grade by several years and the result stops meaning anything.
Gunning Fog Index Calculator
- What is a good Gunning Fog score?
- Aim for 8 to 10 for a wide audience and under 12 for anything meant to be read quickly. Above 17 you are writing for people who are being paid to read it.
- How is the Gunning Fog index calculated?
- Add average words per sentence to the percentage of complex words, then multiply by 0.4. A complex word is three or more syllables, excluding proper nouns, familiar compound words and verb forms made three syllables only by an -ed or -es ending.
- What counts as a complex word?
- Three syllables or more. Gunning's original rules excluded proper nouns, easy compounds like bookkeeper, and words that only reach three syllables through an inflected ending. Automated tools including this one apply the syllable rule directly, so expect a slightly stricter score than a hand count.
- How is Gunning Fog different from Flesch-Kincaid?
- Flesch-Kincaid averages syllables across every word, so many short words dilute a few long ones. Fog counts what share of words are long, so jargon shows up more sharply. For business and technical writing Fog is usually the more revealing of the two.
- Why is my Fog score higher than my Flesch-Kincaid grade?
- Because your text has a concentration of long words rather than a uniform spread. That gap between the two scores is itself the signal: it means jargon, not sentence structure, is what is making the page hard.
- Does the Fog index penalise technical terms unfairly?
- It penalises them accurately - a three-syllable term is harder for a newcomer regardless of how routine it feels to you. Whether that matters depends on your reader. For internal engineering docs, ignore it; for a landing page, do not.
- Who invented the Gunning Fog index?
- Robert Gunning, a US businessman who founded a consultancy in 1944 to fix what he called 'foggy writing' in newspapers and business documents. He published the index in 1952.
- How much text do I need?
- At least 100 words, and ideally several hundred. The complex-word percentage is unstable on short samples - one technical term in a 20-word snippet swings the score by several grades.
QR Code Generator
- Do these QR codes expire?
- No. The code encodes your URL directly, with no redirect service in the middle. It works forever and nobody can track or disable it - unlike most free generators.
- What error correction level should I use?
- M (15%) for screens and print. Use H (30%) only if you are placing a logo over the centre, since that damage has to be recoverable.
- How small can I print it?
- Roughly 2cm square for a short URL at typical scanning distance. Longer URLs need more modules and therefore more space - shorten the link first.
- Why is my QR code so dense?
- Every character adds modules. A 200-character URL produces a much denser grid than a 30-character one. Use a short URL and the lowest error correction that suits the surface.
- What is the difference between a static and a dynamic QR code?
- Static codes contain the destination itself and cannot be changed, but never expire. Dynamic codes contain a short redirect URL you can re-point later - convenient, but the code dies if the provider does or starts charging.
- Can I add a logo to the middle?
- Yes, up to roughly 20% of the area at error correction level H, which reserves enough redundancy to survive the obstruction. Test the result before printing.
- Why will my QR code not scan?
- Usually insufficient contrast, no quiet zone, or printing too small. Dark code on a light background, four modules of blank margin, and at least 2cm square fixes nearly all failures.
- Do QR codes work in print and on screen equally?
- Yes, but print needs more size margin. Screens are backlit and high contrast; a code on matte paper under poor lighting needs to be larger.
Regex Tester
- Which regex flavour is this?
- JavaScript's, via the native RegExp engine. Most syntax is shared with PCRE, but lookbehind, named groups and unicode property escapes differ in support across languages.
- What does the g flag change?
- Without it, only the first match is returned. With it, the engine continues from the end of each match. It is the flag people forget most often.
- Why does my pattern hang the page?
- Catastrophic backtracking - nested quantifiers like (a+)+ against a long non-matching string. We cap the match loop, but the real fix is to rewrite the pattern to avoid ambiguity.
- Do I need to escape forward slashes?
- Not here - you are entering the pattern body without delimiters. In a JavaScript literal you would need to escape them.
- Why does my regex work here but not in Python?
- Flavour differences. JavaScript has no lookbehind in older engines, no atomic groups, and different Unicode property syntax. Test in the flavour you will ship in.
- What is the difference between greedy and lazy matching?
- Greedy quantifiers like .* take as much as possible then backtrack; lazy ones like .*? take as little as possible then expand. For matching inside delimiters, lazy is almost always what you want.
- How do I match across multiple lines?
- The m flag makes ^ and $ match at each line boundary. The s flag makes . match newlines. They are different things and you often need both.
- Should I parse HTML with regex?
- No. HTML is not regular and any pattern that appears to work fails on nested or malformed markup. Use DOMParser in the browser or a real parser on the server.
Base64 Encode / Decode
- Is base64 encryption?
- No. It is an encoding, fully reversible by anyone. It provides zero security. Never use it to hide a secret.
- What is base64url?
- A variant that replaces + with - and / with _ so the output is safe in URLs and filenames. JWTs use it. This tool accepts both on decode.
- Why does my decoded text look wrong?
- Almost always a UTF-8 problem - naive atob returns a binary string and mangles anything outside ASCII. This tool decodes through TextDecoder, so accented and non-Latin characters come back intact.
- How much larger is base64?
- About 33%. Three bytes become four characters, plus padding. That is why inlining large images as data URIs bloats a page.
- Why does base64 end in equals signs?
- Padding. The encoding works in three-byte groups; when the input does not divide evenly, one or two = characters pad the final group. base64url usually omits them.
- When should I use a data URI instead of a file?
- For images under about 2KB used above the fold, where the extra 33% is cheaper than a round trip. Anything larger is better as a real file the browser can cache.
- How do I encode base64 in the terminal?
- base64 < file on macOS and Linux, base64 -d to decode. Add -w 0 on GNU coreutils to stop it wrapping lines.
- Is base64 the same as encryption or hashing?
- No to both. Encryption needs a key and hashing is one-way. Base64 needs neither and reverses in a single step - it hides nothing from anyone.
Image Resizer
- Is my image uploaded anywhere?
- No. Decoding and resizing happen on a canvas in your browser using createImageBitmap. There is no network request, which is why this works offline once the page has loaded.
- Will resizing lose quality?
- Scaling down is close to lossless in perceived quality. Scaling up cannot invent detail - a 200px image enlarged to 800px will look soft no matter which tool you use.
- What size should a hero image be?
- Match the largest size it actually renders at, then serve it as WebP. A 4000px image displayed at 1200px wastes roughly 90% of its bytes.
- Does it keep transparency?
- Yes. Output is PNG, which preserves the alpha channel. Use the WebP converter if you want transparency at a smaller file size.
- What size should images be for a website?
- Match the largest size the image actually renders at, then serve WebP. A full-width hero rarely needs more than 1600px wide; a thumbnail displayed at 300px should not be a 3000px file.
- Does resizing an image reduce its file size?
- Substantially. File size scales roughly with pixel count, so halving both dimensions cuts the data by about 75%.
- What is the difference between resizing and cropping?
- Resizing scales the whole image and keeps everything. Cropping cuts pixels away and changes the composition. Resize to fit a layout, crop to change the framing.
- Will resizing to a different aspect ratio distort the image?
- Yes, if you unlock the ratio. Keep the lock on unless you specifically want the stretch.
Image Compressor
- What quality setting should I use?
- 80% is the usual sweet spot for photographs - typically a 60 to 80% size reduction with no visible difference. Below 60% compression artefacts start showing on gradients and skin tones.
- Is this lossless?
- No. It re-encodes as JPEG, which is lossy by definition. That is what makes the saving large. For lossless compression you want a PNG optimiser instead.
- Why did my PNG get bigger?
- PNGs with few colours are already efficiently compressed, and re-encoding a screenshot as JPEG can grow it. For flat-colour images, convert to WebP instead.
- How small should a web image be?
- Under 200KB for a hero, under 100KB for anything else. Google flags images over roughly 500KB as a Core Web Vitals problem, and they are usually the cause of a slow LCP.
- How small should a web image be?
- Under 200KB for a hero and under 100KB for everything else. Google's Lighthouse flags images that could save more than 4KB, which in practice means most unoptimised photos.
- What is the difference between lossy and lossless compression?
- Lossy discards detail the eye is unlikely to notice and cannot be undone. Lossless rearranges the data so it reconstructs exactly, and saves far less. JPEG and WebP are lossy; PNG is lossless.
- Why did compressing my screenshot make it worse?
- Screenshots are flat colour and sharp text, which JPEG handles badly - it produces ringing around the edges. Keep screenshots as PNG or lossless WebP.
- Does compressing images help SEO?
- Indirectly and reliably. Images are the largest asset on most pages, so they usually determine Largest Contentful Paint, which is a ranking signal.
SVG to PNG
- Why convert an SVG at all?
- Because some places will not take one: email clients, most social preview images, app store assets, and Open Graph tags. SVG is the better format everywhere it is accepted.
- What scale should I export at?
- Work out the pixel size you need and pick the scale that reaches it. An SVG with a 256px viewBox exported at 2x gives 512px, which covers most retina icon uses.
- Does it keep transparency?
- Yes. The PNG keeps the SVG's transparent background - no white box.
- Why does my SVG render wrong?
- Almost always external references. An SVG that links to a webfont or an external image cannot resolve them inside a canvas. Convert text to paths and embed images as data URIs before exporting.
- What resolution should I export at?
- Twice the size the image will display at, so it stays sharp on retina screens. A 200px logo should be exported at 400px.
- Can I convert PNG back to SVG?
- Not meaningfully. Tracing a bitmap produces an approximation with far more paths than a hand-drawn vector. Keep the original SVG.
- Does the PNG keep the transparent background?
- Yes. The alpha channel is preserved, so there is no white box behind the artwork.
- Why does my exported PNG look blurry?
- The export scale is too low, or the SVG has no width and height attributes so it rasterised at a default size. Set explicit dimensions or a viewBox and raise the scale.
Favicon Generator
- Which favicon sizes do I actually need?
- 16x16 and 32x32 for browser tabs, 180x180 for Apple touch icons, and 192x192 plus 512x512 for Android and PWA manifests. Five files covers every real case.
- Do I still need favicon.ico?
- Rarely. Every browser released in the last decade reads PNG favicons from link tags. A root favicon.ico is only worth keeping for very old clients that request it blindly.
- Why does my favicon look muddy at 16px?
- Detail disappears at that size. Favicons that read well are a single letter or one bold shape - a shrunk-down full logo turns to mush.
- Why does the browser still show the old icon?
- Favicons are cached aggressively and ignore normal cache headers. Add a query string to the href, or check in a private window.
- What size should the source image be?
- At least 512 x 512 and square. Anything smaller has to be upscaled for the manifest icons and looks soft on Android home screens.
- Do I need a site.webmanifest?
- Only if you want the site installable or a custom home-screen icon on Android. It is a small JSON file referencing the 192 and 512 icons.
- Should the favicon match my logo exactly?
- Usually not. A full wordmark is illegible at 16px. Use the mark, a single letter, or a simplified version with more contrast.
- Can a favicon be an SVG?
- Yes, and modern browsers prefer it - one file, sharp at every size, and it can respond to dark mode with a media query. Keep PNG fallbacks for Safari and older clients.
PNG to WebP
- How much smaller is WebP?
- Typically 25 to 35% smaller than a comparable JPEG and far smaller than a PNG for photographic content, at visually equivalent quality.
- Is WebP supported everywhere?
- Every current browser supports it, including Safari since version 14. The only holdouts are very old clients, and a picture element with a JPEG fallback covers those.
- Does WebP keep transparency?
- Yes, unlike JPEG. That makes it the best replacement for a PNG logo - transparency preserved, usually a fraction of the size.
- Should I use AVIF instead?
- AVIF compresses better again but encodes slowly and browser support is thinner. WebP is the pragmatic default; ship AVIF as an extra source if you can generate it in your build.
- Should I still keep a PNG fallback?
- No. Every browser released since 2020 supports WebP, including Safari 14 and later. A fallback is now dead weight in your markup.
- Does converting to WebP lose transparency?
- No. WebP supports a full alpha channel, which is what makes it a real replacement for PNG rather than for JPEG only.
- Is WebP lossy or lossless?
- It supports both. This converter uses lossy encoding with a quality slider, which is where the large savings come from. Lossless WebP still beats PNG by about 25%.
- Will WebP hurt my image SEO?
- No. Google Images indexes WebP normally, and the format is Google's own. What matters is alt text, filename and that the image loads fast.
Barcode Generator
- Which barcode format should I use?
- Code 128 for anything internal - it encodes the full ASCII set and is the most widely scanned. EAN-13 and UPC-A are only for retail products with a registered GS1 number.
- Can I make my own EAN-13 number?
- You can generate the image, but a valid retail barcode requires a company prefix from GS1. Inventing digits produces a scannable code that collides with someone else's product.
- How small can I print a barcode?
- Keep the narrowest bar at or above 0.33mm and leave a quiet zone of at least ten times that on each side. Squeezing the quiet zone is the most common reason a printed barcode will not scan.
- Why is my value rejected?
- Each symbology has strict rules - EAN-13 needs exactly 12 or 13 digits, Code 39 accepts only uppercase and a few symbols, Pharmacode only numbers from 3 to 131070.
- What is the difference between a barcode and a QR code?
- A barcode is one-dimensional and holds around 20 characters. A QR code is two-dimensional and holds thousands, including URLs. Use a barcode for inventory numbers, a QR code for links.
- Do I need to buy a barcode for my product?
- For retail, yes - a GS1 company prefix. Retailers reject numbers that do not resolve to a registered owner. For internal warehouse or asset tracking, generate whatever you like.
- What is the check digit?
- The final digit of EAN and UPC codes, computed from the others. It lets a scanner detect a misread. Enter 12 digits for EAN-13 and the check digit is calculated for you.
- What resolution should I export for print?
- At least 300 DPI, and never scale the image after export - resampling blurs the bar edges and scanners start failing.
Code 128 Barcode Generator
- What is Code 128 used for?
- Anything internal: asset tags, warehouse locations, shipping references, work orders. It is not a retail symbology - a supermarket till expects EAN-13 or UPC-A - but for every barcode that never crosses a checkout, Code 128 is the default.
- How many characters can Code 128 hold?
- There is no fixed limit in the spec, but practice caps it around 20 to 25 characters. Past that the symbol gets too wide to scan in one pass and too wide to print on a normal label. If you need more, use a QR code.
- Do I need to add a check digit?
- No. Code 128 mandates a modulo-103 check character and it is computed and appended for you. Unlike EAN, it is not part of the human-readable text, so do not type one in.
- What is the difference between Code 128 and GS1-128?
- GS1-128 is Code 128 with rules on top: an FNC1 start character and standardised Application Identifiers that label each data field, such as (01) for a product number or (17) for an expiry date. The bars are identical; the meaning is not. This tool makes plain Code 128, so do not use it for a GS1-128 logistics label.
- Why is Code 128 narrower than Code 39?
- Subset C packs two digits into every symbol, and Code 128 uses variable-width bars where Code 39 uses a fixed wide/narrow pattern. On a numeric value Code 128 is typically half the width, which matters a lot on a small label.
- Can Code 128 encode lowercase letters?
- Yes, through subset B - that is one of its main advantages over Code 39, which is uppercase only. Case is preserved exactly, so scan-back into a case-sensitive system works.
- How small can I print a Code 128 label?
- Keep the narrowest bar at or above 0.25mm and leave a quiet zone of at least 10 narrow bars on both sides. A squeezed quiet zone is the single most common reason a correctly generated barcode will not scan.
- Will every scanner read Code 128?
- Effectively yes on modern hardware. Some very old laser scanners ship with Code 128 disabled in firmware and need it switched on in the configuration sheet, but nothing made in the last two decades struggles with it.
EAN-13 Barcode Generator
- Can I use a free EAN-13 number?
- You can generate a scannable image from any 12 digits, and for a mockup or an internal test that is fine. For actual retail you cannot: the number must come from a GS1 company prefix, and major retailers verify that a barcode resolves to a registered owner before they will list the product.
- How is the EAN-13 check digit calculated?
- Multiply the digits alternately by 1 and 3 from the left, sum them, and the check digit is whatever brings that sum up to the next multiple of ten. Enter 12 digits here and it is done for you.
- What is the difference between EAN-13 and UPC-A?
- UPC-A is 12 digits and EAN-13 is 13. A UPC-A code is really an EAN-13 with a leading zero, so every EAN-13 scanner reads UPC-A. The reverse is not guaranteed on older North American hardware, which is why products sold in the US usually carry UPC-A.
- Does the first digit tell me the country of origin?
- No, and this is the most persistent barcode myth. The prefix identifies the GS1 member organisation the company registered with, not where the goods were made.
- What if my number is only 8 digits?
- That is EAN-8, a separate short symbology for packaging too small for a full EAN-13. It is issued separately by GS1 and is not a truncated EAN-13.
- How large should an EAN-13 barcode be printed?
- Nominal size is 37.29mm by 25.93mm. The permitted range runs from 80% to 200% of that. Below 80% the bar widths fall under what checkout scanners are calibrated for and read rates collapse.
- Can I put an EAN-13 barcode on a coloured background?
- Only if the contrast holds. Dark bars on a light background is the rule - black on white is safest. Red bars are effectively invisible to the red laser in older scanners, so never print bars in red or place them on a red background.
- Do I need a new barcode for each size or colour?
- Yes. Every variant a retailer prices or stocks separately needs its own item reference. A shirt in three sizes and two colours is six EAN-13 numbers.
UPC Barcode Generator
- What is the difference between UPC and EAN?
- Length. UPC-A is 12 digits, EAN-13 is 13. A UPC-A code is an EAN-13 with a leading zero, so EAN scanners read both. Use UPC-A for North American retail and EAN-13 elsewhere.
- Where do I get a real UPC number?
- From GS1 US, as a company prefix you then assign item references under. Resellers offering cheap single barcodes are selling numbers from prefixes issued before 2002 - they scan, but many large retailers reject them because the registered owner is not you.
- How is the UPC check digit calculated?
- Multiply the digits alternately by 3 and 1 from the left, sum them, and the check digit is what brings the total to the next multiple of ten. Enter 11 digits here and it is computed for you.
- What is UPC-E?
- A compressed 8-digit form for small packaging, produced by suppressing zeros from a UPC-A number. It is not a separate registration - it encodes the same product, just shorter.
- How big does a UPC barcode need to be?
- Nominal is 37.29mm by 25.91mm, and retailers accept 80% to 200% of that. Anything under 80% starts failing at the till.
- Can I shrink the height to fit my label?
- Not much. Truncating bar height is technically scannable but it forces the operator to align the scanner precisely, and most retailer packaging specs prohibit it outright.
- Do I need a separate UPC for each product variant?
- Yes - every size, flavour, colour or multipack a retailer stocks and prices separately needs its own item reference under your company prefix.
- Why does my UPC scan but show the wrong product?
- Almost always because the number belongs to someone else's company prefix - either invented, or bought from a reseller. The barcode is valid; the ownership record is not yours.
Code 39 Barcode Generator
- Should I use Code 39 or Code 128?
- Code 128 unless something specifically demands Code 39. Code 128 handles lowercase, is roughly half the width for the same data, and is read by everything modern. Code 39 survives because of legacy specs - defence logistics, some automotive and healthcare systems - not because it is better.
- Why is my lowercase text rejected?
- Standard Code 39 has no lowercase. There is an extension, Full ASCII Code 39, that encodes lowercase as two-character escape sequences, but scanners must be configured for it and many are not. Uppercase your value, or switch to Code 128.
- Do I need the asterisks?
- They are the start and stop characters and are added for you. If you type them into the value you will produce a barcode with doubled delimiters that most scanners refuse.
- Does Code 39 have a check digit?
- Optionally - a modulo-43 check character. It is not part of the base spec and most implementations omit it, so only add one if the receiving system explicitly asks for it.
- Why is Code 39 so wide?
- Every character is a fixed nine elements, three of them wide, with no compression for digits. A 10-character value produces a noticeably longer symbol than the same value in Code 128 subset C.
- What is the wide-to-narrow ratio?
- Between 2:1 and 3:1, with 3:1 scanning most reliably. Below 2:1 scanners cannot tell wide bars from narrow ones and read rates fall away.
- Can Code 39 encode spaces?
- Yes, space is a valid character in the set. It is one of the few symbologies where a space inside the value is entirely legitimate.
- Is Code 39 the same as Code 3 of 9?
- Yes - two names for the same symbology. The name comes from the encoding: nine elements per character, three of them wide.
ITF-14 Barcode Generator
- What is ITF-14 used for?
- Outer cases and shipping cartons - the box the retail units travel in, not the units themselves. The individual products inside carry EAN-13 or UPC-A; the carton carries ITF-14.
- What are bearer bars and do I need them?
- The thick frame around the symbol. They stop a partial scan being read as a complete one when a scanner clips the edge of the code, and on ITF-14 they are mandatory in the GS1 spec - at minimum as a bar above and below.
- Why must ITF-14 have an even number of digits?
- Interleaved 2 of 5 encodes digits in pairs, one in the bars and one in the spaces. An odd count cannot be paired, which is why the format is fixed at 14 and why a short value gets zero-padded.
- Can I use the same number as my EAN-13?
- Not directly. Take the 13-digit GTIN, add an indicator digit at the front for the pack level, then recalculate the check digit for the new 14-digit number - the old check digit is no longer valid.
- Why is ITF-14 printed so large?
- It is printed directly onto corrugated board, where ink spreads and the surface is uneven. The wide bars and generous size are what make it survive flexographic printing on cardboard.
- Can ITF-14 encode letters?
- No, digits only. If you need text on a carton label use GS1-128, which is the symbology for logistics data such as batch numbers and expiry dates.
- What size should an ITF-14 be?
- Nominal width is 142.75mm at 100%, with a permitted range of 50% to 100% depending on print quality. Direct printing on corrugated needs the larger end; a printed label applied to the carton can go smaller.
- Does every carton need its own barcode?
- Every distinct case configuration does. Identical cartons of the same product share one ITF-14; a 12-pack and a 24-pack of the same item need different indicator digits and therefore different codes.
Morse Code Translator
- How is Morse written down?
- A space separates letters and a slash separates words. Some sources use three spaces for a word break instead - this tool accepts both when decoding.
- Does Morse have upper and lower case?
- No. Morse encodes letters without case, so decoding always returns lowercase. Encoding is case-insensitive.
- What is SOS in Morse?
- ... --- ... It was chosen because the pattern is unmistakable, not as an abbreviation - the popular expansions to Save Our Souls came later.
- Which punctuation is supported?
- Period, comma, question mark, apostrophe, exclamation, slash, brackets, ampersand, colon, semicolon, equals, plus, hyphen, underscore, quote, dollar and at.
- How do I learn Morse code?
- By sound, not by sight. Learn each letter as a rhythm at full speed with longer gaps between them - the Koch and Farnsworth methods - because learning a lookup table caps you at a few words a minute.
- Is Morse code still used?
- Yes. Amateur radio uses it daily, aviation navigation beacons identify themselves in Morse, and it remains a reliable fallback when a signal is too weak for voice.
- What is the timing of Morse code?
- A dash is three dots long, the gap inside a letter is one dot, between letters three, between words seven. That ratio is what makes it readable at any speed.
- Can Morse code encode numbers and punctuation?
- Numbers yes, with five-element patterns. Punctuation is defined for the common marks - period, comma, question mark, slash and a few others - but rarely used on air.
Percentage Calculator
- How do I calculate a percentage increase?
- Subtract the old value from the new, divide by the old value, multiply by 100. Going from 40 to 50 is a 25% increase, because 10 divided by 40 is 0.25.
- Why is a 50% rise then a 50% fall not back to the start?
- Because the second percentage applies to a larger base. 100 up 50% is 150; 150 down 50% is 75. Percentage changes do not cancel out.
- What is the difference between percent and percentage points?
- Moving from 10% to 12% is a rise of 2 percentage points but a 20% increase. Confusing the two is the most common error in reporting statistics.
- How do I remove a percentage from a total?
- Divide, do not subtract. To remove 20% VAT from 120, divide by 1.2 to get 100 - subtracting 20% gives 96, which is wrong.
- How do I calculate a discount?
- Multiply by one minus the discount. 30% off 80 is 80 x 0.7 = 56. To find the original price from a discounted one, divide: 56 / 0.7 = 80.
- How do I work out VAT or sales tax?
- To add 20%, multiply by 1.2. To remove it from a gross figure, divide by 1.2. Subtracting 20% from the gross gives the wrong answer.
- What is the formula for percentage change?
- (new - old) / old x 100. Use the old value as the denominator; using the new one gives a different and usually wrong number.
- How do I calculate a percentage of a percentage?
- Multiply the decimals. 20% of 50% is 0.2 x 0.5 = 0.1, which is 10% of the original.
Roman Numeral Converter
- What is the largest roman numeral?
- MMMCMXCIX, which is 3999. Classical notation has no way to write 4000 or above without the overline convention, and no symbol for zero at all.
- Why is IIII wrong?
- Standard notation uses subtractive pairs - four is IV, not IIII. Clock faces often use IIII for visual balance, which is a design convention rather than correct notation.
- Which subtractive pairs are allowed?
- Only six: IV, IX, XL, XC, CD and CM. You cannot write IL for 49 - it is XLIX.
- How do I read a long numeral?
- Left to right, adding each value, except where a smaller symbol sits before a larger one, in which case subtract it. MCMXCIV is 1000 + (1000−100) + (100−10) + (5−1) = 1994.
- How do you write large numbers in Roman numerals?
- Classical notation stops at 3999 (MMMCMXCIX). Above that, medieval scribes drew a bar over a numeral to multiply it by a thousand, a convention called vinculum, which has no standard plain-text form.
- What year is MMXXVI?
- 2026. M is 1000, so MM is 2000, XX is 20, V is 5 and I is 1.
- Is there a Roman numeral for zero?
- No. The system has no zero and no negative numbers - it was built for counting and tallying, not arithmetic. Latin writers used the word nulla when they needed nothing.
- Why do clock faces use IIII instead of IV?
- Tradition and visual balance. IIII mirrors the VIII opposite it, and the practice predates the standardisation of subtractive notation. It is not an error, but it is not standard notation either.
Unit Converter
- Are the conversion factors exact?
- Yes for defined units. An inch is exactly 25.4mm and a pound is exactly 453.59237g by international agreement, not approximations.
- Is a kilobyte 1000 or 1024 bytes?
- This tool uses 1024, which is what operating systems and developers mean in practice. Storage manufacturers use 1000, which is why a 1TB drive shows as roughly 931GB.
- Why is temperature handled separately?
- Temperature scales have different zero points, so they need offset arithmetic rather than simple multiplication. Everything else here is a ratio conversion.
- US or imperial gallons?
- US. A US gallon is 3.785 litres and an imperial gallon is 4.546 - a 20% difference that catches people out constantly.
- How do I convert Celsius to Fahrenheit?
- Multiply by 9/5 and add 32. Going back, subtract 32 first then multiply by 5/9. Temperature is the one family where you cannot just multiply by a factor, because the scales have different zero points.
- Is a kilobyte 1000 or 1024 bytes?
- Both, depending on who is asking. Operating systems and developers use 1024. Storage manufacturers use 1000, which is why a 1TB drive shows as 931GB. The unambiguous names for the binary units are kibibyte and mebibyte.
- How many millilitres in a cup?
- 236.6 in the US, 250 in the metric cup used in Australia and much of Europe, and 284 for an imperial cup. Recipes rarely say which, which is why baking by weight is more reliable.
- Are US and imperial measurements the same?
- Not for volume. A US gallon is 3.785 litres and an imperial gallon is 4.546 - about 20% larger. Length and weight units do match.
WebP to PNG
- Why convert WebP to PNG at all?
- Because some software still refuses WebP - older design tools, some CMS uploaders, certain print workflows and a few email clients. PNG is accepted everywhere.
- Will the PNG be larger?
- Almost always, often several times. PNG is lossless and WebP is usually lossy, so you are unpacking compressed data. Convert for compatibility, not for quality.
- Does transparency survive?
- Yes. Both formats support an alpha channel, so a transparent WebP produces a transparent PNG.
- Can I recover quality lost by WebP compression?
- No. Detail discarded during lossy compression is gone. Converting to a lossless format preserves what remains; it cannot restore what was thrown away.
- Why will nothing open my WebP file?
- It is a relatively recent format and some desktop software never added support. Converting to PNG is the reliable fix; PNG has been universal for thirty years.
- Will the PNG be bigger than the WebP?
- Nearly always, often several times over, because PNG is lossless and the WebP was probably lossy. That is expected, not a fault.
- Can converting improve a blurry WebP?
- No. Detail discarded by lossy compression is gone permanently. Converting preserves what is left; it cannot reconstruct what was thrown away.
- How do I save a WebP image from a website as PNG?
- Download the WebP, then convert it here. Right-click save-as in some browsers already offers PNG, but it re-encodes the same way this does.
Epoch Converter
- Seconds or milliseconds?
- Unix time is defined in seconds. JavaScript's Date.now returns milliseconds. A 10-digit number is seconds, a 13-digit number is milliseconds - mixing them is the single most common date bug there is.
- What is the year 2038 problem?
- A signed 32-bit timestamp overflows on 19 January 2038. Anything storing time in a 32-bit integer breaks then. Use 64-bit integers or a native timestamp type.
- Does Unix time account for leap seconds?
- No. It assumes every day is exactly 86,400 seconds, so it drifts from actual astronomical time. That is deliberate - it keeps the arithmetic simple.
- Why does my date show a day off?
- Almost always a timezone issue. A date with no time component is midnight UTC, which is the previous day in any negative offset. Store and compare in UTC.
- Is my timestamp in seconds or milliseconds?
- Count the digits. Ten digits is seconds and lands in the present; thirteen is milliseconds. A ten-digit value read as milliseconds gives a date in 1970, which is the classic symptom.
- What is the current Unix timestamp?
- It increments every second. Get it with Math.floor(Date.now()/1000) in JavaScript, time.time() in Python, or date +%s in a shell.
- Why does my timestamp show the wrong day?
- Timezone. A Unix timestamp is always UTC; the date you see depends on the zone used to format it. A midnight UTC timestamp is the previous evening in the Americas.
- Should I store dates as timestamps or as dates?
- Store an instant as timestamptz in Postgres, which is a timestamp with the zone handled correctly. Store a calendar date with no time - a birthday, an invoice date - as a plain date, never as an epoch.
Caesar Cipher
- What is ROT13?
- A Caesar cipher with a shift of 13. Because the alphabet has 26 letters, applying it twice returns the original - which is why it is used to hide spoilers rather than secrets.
- How do I crack a Caesar cipher?
- Try all 25 shifts and read which one produces words. That is why this tool lists them all - with only 25 possibilities there is nothing to break.
- Is a Caesar cipher secure?
- Not remotely. It has 25 possible keys and falls instantly to frequency analysis. It is a teaching tool and a party trick, never a security measure.
- What happens to numbers and punctuation?
- They pass through unchanged. Only the 26 letters are shifted, and case is preserved.
- How do I decode a Caesar cipher without the key?
- Try all 25 shifts and read which produces words - a brute force that takes seconds. For longer texts, letter frequency alone reveals the shift immediately.
- What is the difference between a Caesar cipher and ROT13?
- ROT13 is a Caesar cipher with the shift fixed at 13. Because 13 is half of 26, encoding and decoding are the same operation.
- Why is a Caesar cipher not secure?
- The key space is 25. Any computer tries every option instantly, and even by hand frequency analysis breaks it in minutes. It was adequate when most opponents could not read at all.
- What is a Vigenere cipher?
- A Caesar cipher with a shift that changes per letter according to a keyword. It resisted analysis for three centuries and is still trivially broken today, but it is the natural next step up.
Image Color Picker
- Is my image uploaded?
- No. It is drawn to a canvas in your browser and read back with getImageData. There is no network request at any point.
- Why does the picked colour look different from the original?
- Screenshots and photographs are often in a wider colour profile than sRGB. The canvas reads sRGB values, which can differ slightly from what a design tool reports.
- When should I use HSL over HEX?
- HSL when you are building a palette. Changing only the lightness gives you consistent tints and shades of the same hue, which is very difficult to do by hand in HEX.
- Can I pick from a screenshot?
- Yes - take the screenshot, drop it in, and click. That is the usual way to lift a colour from a site or app you cannot inspect.
- How do I find the hex code of a colour in a screenshot?
- Drop the screenshot in and click the pixel. That is the reliable way to lift a brand colour from a site or app whose CSS you cannot inspect.
- What is the difference between HEX, RGB and HSL?
- They describe the same colour three ways. HEX and RGB are the same numbers in different notation. HSL splits it into hue, saturation and lightness, which makes building tints and shades far easier.
- Why does the colour differ from the design file?
- Colour profile. Photos and screenshots are often in Display P3 or Adobe RGB; the canvas reads sRGB. The values are correct for the web, which is sRGB.
- Can I pick an average colour rather than one pixel?
- Zoom out and pick from a flat area. A single pixel in a gradient or a JPEG artefact is not representative of the region it sits in.
Subnet Calculator
- Why are two addresses unusable?
- The first is the network address and the last is the broadcast address. Neither can be assigned to a host, which is why a /24 gives 254 usable addresses rather than 256.
- What about /31 and /32?
- A /32 is a single host route. A /31 is a special case defined for point-to-point links where both addresses are usable, since there is no need for a broadcast address.
- What is a wildcard mask?
- The bitwise inverse of the subnet mask. Cisco ACLs and OSPF use it instead of a subnet mask, so /24 is 0.0.0.255 rather than 255.255.255.0.
- Which ranges are private?
- 10.0.0.0/8, 172.16.0.0/12 and 192.168.0.0/16 are private. 100.64.0.0/10 is carrier-grade NAT, and 169.254.0.0/16 is link-local.
- How many hosts are in a /24?
- 254 usable. A /24 spans 256 addresses, minus the network address and the broadcast address. The general formula is 2^(32-prefix) - 2.
- What does /24 actually mean?
- The first 24 bits are the network portion, leaving 8 bits for hosts. It is the same information as the mask 255.255.255.0, written more compactly.
- How do I split a network into smaller subnets?
- Borrow bits from the host portion. Splitting a /24 into four means moving to /26, giving 64 addresses and 62 usable hosts each.
- How is IPv6 subnetting different?
- There is no broadcast address and no scarcity. A /64 is the standard size for any single network regardless of how many devices it holds, and you subnet by taking prefixes shorter than /64.
Wildcard Mask Calculator
- What is a wildcard mask?
- The bitwise inverse of a subnet mask, used by Cisco IOS to say which address bits matter. A 0 means the bit must match exactly; a 1 means ignore it. It is the opposite convention to a subnet mask, which is why they are so easy to mix up.
- How do I convert a subnet mask to a wildcard mask?
- Subtract each octet from 255. 255.255.255.0 becomes 0.0.0.255; 255.255.240.0 becomes 0.0.15.255. That is all there is to it.
- Where do I need a wildcard mask instead of a subnet mask?
- Cisco IOS access control lists, OSPF and EIGRP network statements, and route maps. Interface addressing still takes a normal subnet mask, which is the source of most of the confusion.
- What does a wildcard mask of 0.0.0.0 mean?
- Match this exact address and nothing else - the equivalent of a /32. In an ACL you can write it as the keyword host instead, and most style guides prefer that for readability.
- What does 255.255.255.255 mean as a wildcard mask?
- Ignore every bit, so match any address at all. IOS lets you write this as the keyword any.
- Can a wildcard mask be non-contiguous?
- Yes, and that is the one real capability a subnet mask lacks. A wildcard of 0.0.0.254 matches only even-numbered hosts. It is legal, occasionally useful, and worth a comment in the config because the next engineer will assume it is a typo.
- Why does Cisco use wildcard masks at all?
- They came from the original ACL implementation, where inverting the mask made the bit test cheaper in hardware. The convention outlived the reason, and newer platforms such as NX-OS and ASA accept prefix lengths directly.
- Is a wildcard mask the same as an inverse mask?
- Yes - inverse mask, wildcard mask and, in older Cisco documentation, don't-care bits all describe the same thing.
PNG to PDF
- How do I convert a PNG to PDF without uploading it?
- Use a converter that runs in the browser, like this one. The image is decoded on a canvas and the PDF is assembled locally, so the file never reaches a server.
- Will the PDF have a watermark?
- No. Nothing is added to the page - the PDF contains your image and nothing else.
- What page size does the PDF use?
- The page matches the image's pixel dimensions at 72 points per inch, so the image fills the page exactly with no borders or cropping.
- Can I put several PNGs into one PDF?
- Not here - each conversion produces a single-page PDF. To combine them, convert each and merge the PDFs in Preview on macOS or any PDF merger.
- Does converting reduce the image quality?
- Slightly. The image is stored as a JPEG stream inside the PDF, which is what keeps the file small. Raise the quality slider if you need a near-lossless result.
- Why is my PDF so large?
- Because a photograph at full resolution is large. Lower the quality slider, or resize the image first - a 4000px wide image is far more than a printed page needs.
- Does transparency survive the conversion?
- No. PDF pages are opaque, so transparent areas are filled with white. Keep the PNG if you need the alpha channel.
- Can I convert a PDF back to PNG?
- Yes - use the PDF to PNG converter, which rasterises every page at twice its natural size.
PDF to PNG
- How do I convert a PDF to an image without uploading it?
- Use a browser-based converter like this one. The PDF is parsed and rendered on a canvas in the page, so a confidential document never reaches a server.
- What resolution are the PNGs?
- Twice each page's natural size, which is roughly 144 DPI for a standard document. That is sharp enough for screens and for pasting into slides.
- Can I convert only one page?
- Yes. Every page is previewed and each has its own download link, so you can take just the one you need.
- Why did my PDF fail to render?
- Password-protected PDFs cannot be opened without the password, and a small number of files with unusual embedded fonts render imperfectly. Try re-saving the PDF from its source application.
- Should I use PNG or JPG for PDF pages?
- PNG for anything with text, diagrams or line art, because it is lossless and keeps edges crisp. JPG only for pages that are entirely photographic.
- Is there a page limit?
- No hard limit, but a very long document takes time and memory because every page is rendered at once. Split anything over about fifty pages.
- Does this work offline?
- Once the page has loaded, yes. Rendering happens entirely in your browser with no further network requests.
- How is this different from taking a screenshot?
- A screenshot captures your screen resolution and whatever is visible. This renders the page from the PDF's own vector data at double size, so text stays sharp when you zoom in.
PNG to JPG
- Why convert PNG to JPG?
- File size. A photograph saved as PNG is often five to ten times larger than the same image as JPEG at 80% quality, with no visible difference.
- Does converting to JPG lose transparency?
- Yes. JPEG has no alpha channel, so transparent areas become white. Convert to WebP instead if you need to keep transparency and still save space.
- What quality setting should I use?
- 80% for photographs. Below 60% compression artefacts become visible around edges; above 90% the file grows quickly for very little gain.
- Should screenshots be converted to JPG?
- No. JPEG handles flat colour and sharp text badly, producing halos around letters. Keep screenshots as PNG or convert them to lossless WebP.
- Is JPG or JPEG correct?
- They are the same format. The three-letter extension is a leftover from MS-DOS filename limits; both are read identically everywhere.
- Can I convert back to PNG afterwards?
- You can, but the discarded detail does not come back. The result is a lossless copy of a lossy image and will be larger than the JPG with no gain in quality.
- Is my image uploaded anywhere?
- No. Decoding and re-encoding happen on a canvas in your browser. There is no upload and no server involved.
- Does converting strip EXIF data?
- Yes. Canvas re-encoding drops all metadata, including camera details and GPS coordinates, which is useful before publishing a photo.
JPG to PNG
- Why convert JPG to PNG?
- Because you are about to edit it. Every JPEG save re-compresses and loses more detail; working in PNG means no further degradation. Some tools and CMS uploads also reject JPEG.
- Will the PNG be bigger than the JPG?
- Almost always, often several times over. PNG is lossless and stores every pixel; JPEG throws detail away. The size increase is expected.
- Does converting improve image quality?
- No. Detail lost to JPEG compression is gone permanently. Converting preserves exactly what remains, which is all any format can do.
- How do I make the background transparent?
- Converting does not add transparency - it only preserves it. You need an editor or a background-removal tool to cut the subject out first.
- Does the conversion remove EXIF data?
- Yes. Redrawing on a canvas drops camera, date and GPS metadata, which is a useful side effect before sharing a photo publicly.
- Is PNG or WebP the better choice?
- WebP, if the software you are handing it to accepts it. Lossless WebP is about 25% smaller than PNG for the same pixels.
- Is my file uploaded?
- No. Everything happens in your browser, which is why this works on a phone with a poor connection.
- Does this handle HEIC photos from an iPhone?
- Only if your browser can decode HEIC, which Safari can and most others cannot. Export as JPEG from the Photos app first.
Character Counter
- Do spaces count as characters?
- In almost every limit that matters, yes. X, meta descriptions, SMS and most form validations count spaces. Some academic word limits do not, which is why both numbers are shown.
- What are the common character limits?
- 280 for an X post, 150 to 160 for a meta description, 60 for a title tag, 160 for an SMS segment, 2,200 for an Instagram caption and 3,000 for a LinkedIn post.
- How many characters is 500 words?
- Roughly 2,500 to 3,000 including spaces, since the average English word is about five characters plus a space. It varies with vocabulary.
- Does an emoji count as one character?
- Rarely. Most emoji are two UTF-16 units, and ones with skin tone or family modifiers can be seven or more. This counter reports what JavaScript counts, which is what most web forms count.
- How long should a meta description be?
- Under 160 characters, ideally 120 to 155. Google truncates around 920 pixels rather than a fixed count, so shorter is safer on mobile.
- Is my text stored anywhere?
- No. Counting happens in JavaScript on this page, with no network request, so it also works offline once loaded.
- What counts as a sentence?
- A run of text ending in a full stop, question mark or exclamation mark. Abbreviations with full stops will inflate the count slightly.
- How do I count characters in Excel or Google Sheets?
- Use LEN(A1) for the total, and LEN(SUBSTITUTE(A1," ","")) for the count without spaces.
Case Converter
- How do I change uppercase to lowercase without retyping?
- Paste the text here and pick lowercase. In Word or Google Docs, Shift+F3 cycles through cases for the selected text.
- What is the difference between Title Case and Sentence case?
- Title Case capitalises every word. Sentence case capitalises only the first word of each sentence and any proper nouns. Headlines are usually one or the other consistently, never both.
- When should I use camelCase versus snake_case?
- Follow the language. JavaScript and Java use camelCase for variables, Python and Ruby use snake_case, C# and Go use PascalCase for exported names. URLs and CSS classes use kebab-case.
- Why does camelCase strip my punctuation?
- Because identifiers cannot contain it. Spaces and symbols become word boundaries and are removed, which is exactly what a code identifier needs.
- Does Title Case capitalise small words like of and the?
- Style guides disagree. AP and Chicago leave articles, short prepositions and conjunctions lowercase unless they start the title. This tool capitalises every word, so adjust the small ones by hand if your style guide cares.
- Is my text sent anywhere?
- No. Every conversion runs in JavaScript on this page with no network request.
- How do I convert case in Excel?
- UPPER(A1), LOWER(A1) and PROPER(A1) for Title Case. Excel has no built-in Sentence case function.
- What is aLtErNaTiNg case for?
- Almost entirely for mocking something on the internet. It is included because people ask for it.
Diff Checker
- How do I compare two documents for differences?
- Paste both versions here. For Word documents, use Review then Compare, which understands formatting changes as well as text.
- Is my text uploaded to compare it?
- No. The comparison runs in JavaScript on this page, which matters when the two versions are contract drafts or config files.
- What does the diff algorithm actually do?
- It finds the longest sequence of lines common to both texts and treats everything else as an insertion or deletion. That is why inserting one line at the top does not mark the whole file as changed.
- Can it compare word by word instead of line by line?
- This one works on lines, which is the right granularity for code, config and structured text. For prose where you want word-level detail, paste one sentence per line first.
- How do I diff two files from the terminal?
- diff -u old new for a unified diff, or git diff --no-index old new to get git's colouring and formatting on files outside a repository.
- Why do two identical-looking lines show as different?
- Almost always trailing whitespace or line endings - a file saved on Windows uses CRLF and one saved on macOS uses LF. Both are invisible and both count.
- How large a text can it handle?
- Comfortably a few thousand lines. The comparison is quadratic in the number of lines, so very large files will slow the page down.
- Can I compare JSON?
- Yes, but format both sides first so keys line up on their own lines. Comparing minified JSON produces one enormous changed line.
Aspect Ratio Calculator
- What is 16:9 in pixels?
- Any pair where width divided by height is 1.778 - 1280x720, 1920x1080, 2560x1440 and 3840x2160 are all 16:9. The ratio does not specify a size.
- How do I calculate an aspect ratio from two dimensions?
- Divide both by their greatest common divisor. 1920 and 1080 share a divisor of 120, giving 16:9.
- What aspect ratio should a YouTube thumbnail be?
- 16:9, at 1280x720. Instagram posts are 1:1 or 4:5, Stories and Reels are 9:16, and an Open Graph image is 1.91:1 at 1200x630.
- What is the difference between aspect ratio and resolution?
- Ratio is the shape; resolution is the number of pixels. 1280x720 and 3840x2160 have identical shapes and very different sharpness.
- How do I keep an aspect ratio in CSS?
- Use the aspect-ratio property: aspect-ratio: 16 / 9. It is supported everywhere now and replaces the old padding-top percentage hack.
- What is 21:9 actually?
- A marketing name for ultrawide monitors, which are usually 2560x1080 (64:27) or 3440x1440 (43:18). Neither is exactly 21:9.
- Why does my video have black bars?
- The video's ratio does not match the display's, so the player letterboxes it rather than cropping or stretching. Matching the ratio at export removes them.
- How many megapixels is 4K?
- About 8.3 for 3840x2160. 1080p is 2.1 and 8K is 33.2.
Color Converter
- How do I convert HEX to RGB by hand?
- Split the six digits into three pairs and read each as base 16. 25 is 37, 63 is 99, eb is 235.
- What does the third pair of a hex colour mean?
- Blue. The six digits are red, green and blue, two hex digits each, giving 256 levels per channel and 16.7 million colours.
- What is an eight-digit hex colour?
- The last pair is alpha - opacity from 00 to ff. #2563eb80 is that blue at roughly 50% opacity.
- When is HSL more useful than HEX?
- When building a palette. Keeping hue and saturation fixed and varying only lightness gives consistent tints and shades, which is nearly impossible to do by eye in HEX.
- What contrast ratio do I need?
- 4.5:1 for body text and 3:1 for large text or UI components, under WCAG AA. AAA asks for 7:1, which is difficult with brand colours.
- Do three-digit hex codes work?
- Yes. #f00 expands to #ff0000 - each digit is doubled. It only reaches 4,096 colours, so it is a shorthand for simple values.
- Why does my colour look different in a design tool?
- Colour profile. Design tools often work in Display P3 or Adobe RGB; the web is sRGB. The hex value is the same but the rendered result differs.
- Is RGB the same as RGBA?
- RGBA adds an alpha channel from 0 to 1. Modern CSS also accepts rgb(37 99 235 / 50%), which does the same thing.
CSV to JSON
- How do I convert a CSV to JSON without a script?
- Paste it here. For repeated or very large conversions, use a script instead - Python's csv.DictReader and json.dump are four lines.
- Does it handle commas inside quoted fields?
- Yes. A field wrapped in double quotes may contain commas, and a doubled quote inside it is treated as a literal quote character, following RFC 4180.
- Why are my numbers turning into strings, or vice versa?
- Numeric-looking cells are converted to JSON numbers. That breaks zip codes and phone numbers with leading zeros, so quote those in the source CSV if the leading zero matters.
- What delimiter should I use for European spreadsheets?
- Semicolon. Locales that use a comma as a decimal separator export CSV with semicolons, which is why a French or German export looks like one column in an English tool.
- Is my data uploaded?
- No. Parsing runs in JavaScript on this page, which matters when the file is a customer export.
- What happens to an empty cell?
- It becomes an empty string, not null. Post-process if your schema needs nulls.
- How do I handle nested data?
- CSV is flat and cannot express nesting. Convert first, then reshape the flat objects in code - dotted column names like user.name need a second pass.
- Can I go the other way?
- Yes - the JSON to CSV converter takes an array of objects and produces the header row and rows back.
JSON to CSV
- What JSON shape does this expect?
- An array of flat objects, which is what almost every API returns for a list. A single object is treated as a one-row table.
- What happens to nested objects and arrays?
- They are stringified into one cell. CSV has no way to express nesting, so flatten the data before converting if you need the inner values in their own columns.
- Why do some rows have empty cells?
- The header row is the union of every key across all objects. Any object missing a key gets an empty cell there, which is the correct flat representation.
- How do I open the CSV in Excel without mangling it?
- Use Data then From Text/CSV rather than double-clicking, and set the encoding to UTF-8. Double-clicking often guesses the delimiter and encoding wrongly.
- Are quotes and commas escaped properly?
- Yes. Any value containing a comma, quote or newline is wrapped in double quotes with internal quotes doubled, per RFC 4180.
- Should I use a comma or a semicolon?
- Comma for anything programmatic. Semicolon if the file will be opened in a spreadsheet set to a locale that uses a comma as the decimal separator.
- Is my JSON sent to a server?
- No. Parsing and conversion happen in your browser with no network request.
- How do I do this in a script?
- In Python, pandas.read_json then to_csv is one line each. In Node, the json2csv package or a short reduce over Object.keys does the same.
Password Generator
- How long should a password be?
- At least 16 characters for anything that matters, and 20 or more for a password manager master key or an email account. Length beats complexity: a long lowercase password is stronger than a short one full of symbols.
- Are these passwords safe to use?
- They come from crypto.getRandomValues, which is a cryptographically secure source, and they are generated locally and never sent anywhere. The page also works offline once loaded, which you can verify in the network tab.
- What is entropy and what number should I want?
- Entropy measures how many guesses an attacker needs, in bits. Under 60 bits is weak, 80 is reasonable, and above 100 is beyond any realistic offline attack.
- Should passwords contain symbols?
- They help, but far less than length. Adding four characters to a lowercase-only password buys more entropy than adding one symbol to a short one.
- Why are some characters missing from the pool?
- Look-alikes - l, I, 1, O and 0 - are excluded, because passwords do sometimes get read aloud or copied by hand and those cause real errors.
- Should I change my passwords regularly?
- No. NIST dropped that advice in 2017. Forced rotation makes people pick weaker, more predictable passwords. Change one when there is a reason to.
- What is a passphrase and is it better?
- Several random words - correct horse battery staple. Four random words from a large list is around 50 bits, which is memorable but weaker than a 20-character random password. Use passphrases for what you must memorise and random strings for everything a manager holds.
- Do I still need two-factor authentication?
- Yes. A strong password does nothing against a phishing page or a database breach. Use an authenticator app or a passkey rather than SMS.
Lorem Ipsum Generator
- What does lorem ipsum actually mean?
- Nothing coherent. It is a corrupted extract from Cicero's De Finibus Bonorum et Malorum, written around 45 BC, with words truncated and shuffled until the meaning disappeared.
- Why use placeholder text at all?
- Real copy pulls attention to the words. Placeholder text lets you judge line length, leading and hierarchy on their own, which is why designers have used it for five centuries.
- Should I use lorem ipsum in a client mockup?
- Carefully. It hides the fact that real headlines are longer and real body copy is shorter, and clients often cannot see past it. Use real or realistic copy for anything you are presenting for approval.
- How many words are in a paragraph?
- Here, four to six sentences of eight to twenty words, so roughly 60 to 130 words - close to a real body paragraph.
- Will lorem ipsum hurt my SEO?
- Only if it ships. A live page of placeholder text is thin content and looks abandoned. It is harmless in a mockup and embarrassing in production, which happens more often than anyone admits.
- Is there an alternative to lorem ipsum?
- Plenty, from Bacon Ipsum to Hipster Ipsum. They are more memorable and more distracting, which defeats most of the point.
- How do I generate lorem ipsum in my editor?
- VS Code has an Emmet abbreviation: type lorem20 and press Tab for twenty words. Most editors have something equivalent.
- Is the text the same every time?
- No. Each generation shuffles a fresh selection, so two blocks will not look identical in a mockup.
Minecraft Circle Generator
- How do you make a perfect circle in Minecraft?
- You cannot, strictly. Blocks are square, so every circle is an approximation. The standard method fills a block when the centre of that block falls inside the true circle, which is what this generator does and what every large build uses.
- Should I use an odd or even diameter?
- Odd diameters give you a single centre block, which makes it far easier to align a beacon, a pillar or a nether portal. Even diameters leave a 2x2 centre and are harder to build symmetrically.
- What is the difference between a thin and a thick circle?
- A thin circle marks a block when any of its four edge neighbours is outside the circle. A thick circle also checks the four diagonal neighbours, so the ring is wider and reads more clearly from a distance or from the ground.
- How do I build a sphere from this?
- A sphere is a stack of discs. Build the widest circle at the middle layer, then work up and down using the diameter for each layer, which shrinks as you move away from the centre. Layer n of a sphere of radius r has diameter 2 x sqrt(r squared - n squared).
- What is the biggest circle worth building by hand?
- Past about 60 blocks across the outline stops reading as a circle from ground level and the block count runs into the hundreds. Above that, build it in sections and check the shape from the air.
- Does this work for Bedrock and Java?
- Yes. The grid is a geometry problem, not a game feature, so the same pattern works in every edition and in other voxel games.
- How many blocks do I need?
- The tool counts them for you as you change the diameter. As a rough guide a thin outline uses about 2.8 blocks per unit of diameter - a little under pi, because the outline steps diagonally rather than tracing every edge.
- Can I copy the pattern out?
- Yes. Copy as text gives you the grid as hashes and dots, which pastes into any notes app or a chat window so you can keep it on a second screen while you build.
Color Mixer
- What colors make brown?
- Red, yellow and blue together, or any complementary pair: orange and blue, purple and yellow, or green and red. Start with orange and add blue a little at a time - it is the easiest of the three to control.
- What colors make green?
- Blue and yellow. More yellow moves it towards lime and spring green, more blue towards teal and forest green. Adding a touch of red mutes it into olive or sage.
- What colors make orange?
- Red and yellow. Equal parts give a true orange; more yellow gives amber and more red gives scarlet. Adding a little blue dulls it towards terracotta.
- What colors make purple?
- Red and blue. The result depends heavily on which red and which blue - a warm red gives magenta and a cool blue gives indigo. A touch of white lifts it to lavender.
- What colors make pink?
- Red and white. Pink is a tint rather than a mix, so the more white you add the paler it gets. A speck of blue makes it cooler, a speck of yellow makes it coral.
- Why does blue and yellow make green in paint but not on screen?
- Paint is subtractive: each pigment absorbs wavelengths, and what is left over is what you see. Screens are additive: they emit light, so red and green light make yellow. Same colours, opposite arithmetic.
- What are the primary colors?
- For paint and pigment, red, yellow and blue - they cannot be mixed from anything else. For printing it is cyan, magenta and yellow, and for screens it is red, green and blue. The right set depends on the medium.
- What is a complementary color?
- The colour opposite on the wheel: red and green, blue and orange, yellow and purple. Placed side by side they intensify each other; mixed together they cancel into brown or grey.
Random Letter Generator
- How do I generate a single random letter?
- Set the count to one and generate. Each of the 26 letters has the same chance of appearing, roughly 3.85% each.
- Why did the same letter appear twice?
- Because true randomness has no memory. With repeats allowed, two identical letters in a row is expected about once every 26 draws. Switch off repeats if you need all of them different.
- How many letters can I generate without repeats?
- Twenty-six, one for each letter. Restricted to vowels the ceiling is five, and to consonants it is twenty-one.
- Is this good enough for a password?
- No. Use a password generator instead - it mixes cases, digits and symbols and uses your browser's cryptographic random source. Letters alone are far too weak.
- How is the randomness generated?
- In your browser, from JavaScript's built-in random number source. Nothing is sent anywhere and no result is stored or logged.
- Can I use this for Scattergories or Scrabble?
- Yes, that is the common use. For Scattergories most house rules drop the awkward letters - switch to consonants and re-roll if you land on Q, X or Z.
- What about a random letter for each student?
- Set the count to the size of the class and turn repeats off, so every student gets a different letter. Above 26 you will need repeats back on.
Random Number Generator
- How do I pick a random number between 1 and 100?
- That is the default. Press generate and you get one number, with every value from 1 to 100 equally likely.
- What is the difference between repeats on and off?
- With repeats on, each draw is independent and the same number can come up twice - like rolling a die. With repeats off, numbers are drawn without replacement, like pulling raffle tickets from a hat.
- Is this random enough to pick a giveaway winner?
- Yes for an informal draw. For anything with legal or prize-money weight, use a service that publishes a verifiable seed, so entrants can check the draw was not re-run.
- Are the numbers truly random?
- They are pseudorandom: generated by an algorithm in your browser that is statistically random but not physically so. For games, classrooms and giveaways this is indistinguishable from true randomness.
- Can I use negative numbers?
- Yes. Any whole numbers work as the minimum and maximum, including negatives, and they can be entered in either order.
- How many numbers can I draw at once?
- Up to 1000 with repeats on. With repeats off the limit is the size of the range, since you cannot draw 50 unique numbers from a range of 10.
- Does anything get sent to a server?
- No. The numbers are generated in your browser and nothing about the draw is transmitted, logged or stored.
- Why are the results sorted?
- Only when repeats are off, because that mode is for picking a set rather than a sequence and a sorted set is easier to check against a list. With repeats on the draw order is preserved.
Random Group Generator
- How do I split a class into groups?
- Paste the register one name per line, choose number of groups, and set how many you want. Sizes come out within one person of each other automatically.
- What if the number does not divide evenly?
- Names are dealt one at a time around the groups, so the remainder is spread out. Thirteen people into four groups gives you 4, 3, 3, 3 rather than 4, 4, 4, 1.
- Is it really random?
- Yes. The list is shuffled with a Fisher-Yates shuffle, which gives every possible arrangement an equal chance. Re-shuffling gives a genuinely different result, not a rotation of the last one.
- Can I keep two people apart, or together?
- Not automatically. Shuffle, and if a pairing does not work, shuffle again - with more than a handful of names you will get a different arrangement every time.
- Can I paste from a spreadsheet?
- Yes. Paste a column straight in, or a comma-separated list - both are split correctly, and blank lines are ignored.
- Is my list uploaded anywhere?
- No. The names stay in your browser, are never sent to a server and are gone when you close the tab.
- How many names can it handle?
- There is no set limit. Several hundred names paste and shuffle without any noticeable delay.
Pictionary Word Generator
- What are good Pictionary words?
- Concrete nouns with a distinctive silhouette work best - a lighthouse, a kangaroo, a parachute. Words that only differ from another word by detail, like cup versus mug, are frustrating rather than difficult.
- What is the difference between Pictionary and charades?
- Pictionary words are drawn, so they favour things with a clear shape. Charades are acted, so they favour actions and situations. Switching modes here changes the word list to match.
- How long should each round be?
- Sixty seconds is the standard. Ninety works better for hard words and for younger players; thirty turns the game into a sprint and rewards fast drawers over clear ones.
- How many words do I need for a game?
- Roughly one per player per round. Six players over five rounds is thirty words, so draw a fresh set each round rather than one long list.
- Can I use these for charades at a party?
- Yes. Switch to Charades and the list becomes actions and situations - proposing marriage, arguing with a satnav - that work when acted out rather than drawn.
- Are the words suitable for children?
- The easy level is written for children: everyday objects, nothing abstract and nothing that needs reading skill. Medium and hard assume adult general knowledge.
- Can I get the same word twice?
- Not within one draw - each set is unique. Across separate draws a word can reappear, since the list is shuffled fresh each time.
- Do I need an account?
- No. It runs in your browser, there is no signup, and nothing you generate is stored or sent anywhere.
Valorant Crosshair Generator
- How do I import a crosshair code in Valorant?
- Open Settings, go to the Crosshair tab, choose Import Profile Code, and paste. The game applies it to the profile you currently have selected, so switch profile first if you want to keep your existing one.
- What is the best crosshair in Valorant?
- There is no single best one, but the pattern most professionals share is a short thin inner cross, outlines on, outer lines off, and no centre dot. It covers the least head area at range while still being visible against bright walls.
- Should I use a centre dot?
- It helps with precise placement on flick-heavy agents and gets in the way when spraying. Try size 1 with a short cross - a large dot hides exactly the pixels you are trying to hit.
- What colour crosshair should I use?
- Cyan and green are the common picks because almost nothing on Valorant's maps is those colours. White disappears against bright walls and red disappears against several agent abilities.
- Do outlines slow you down?
- No. They cost nothing in performance and make the crosshair readable against light backgrounds, which is why nearly everyone leaves them on.
- What are outer lines for?
- They show recoil and movement error as they expand. Most players disable them because the extra motion is distracting and the information is available from the spray pattern anyway.
- Does the crosshair affect my accuracy?
- Only through visibility. Valorant does not change hit registration based on the crosshair - a smaller one simply lets you see the target you are aiming at.
- Can I use this for other games?
- The preview is useful for judging any crosshair shape, but the profile code format is specific to Valorant and will not import anywhere else.
Reaction Time Test
- What is a good reaction time?
- Under 250ms is good and under 200ms is genuinely fast. The median adult clicks at roughly 273ms on a simple visual test, so beating 250 puts you ahead of most people who try it.
- What is the average human reaction time?
- About 273ms for a simple visual cue clicked with a mouse. Sound cues are faster, around 170ms, and touch is faster still, because audio and touch signals reach the brain through a shorter path than vision.
- What is the fastest possible reaction time?
- Around 100 to 120ms for a trained person on good hardware. Below that you are anticipating rather than reacting: the retina alone takes about 20 to 40ms to send the signal, and the muscle takes another 30 to 50ms to move the finger.
- Why is my reaction time worse than usual?
- Sleep debt, alcohol, a busy browser tab and a wireless mouse all cost 10 to 50ms each. Age matters too - reaction time is fastest in the early twenties and slows by roughly 2 to 3ms per year after about thirty.
- Does the test measure my monitor as well as me?
- Yes, and there is no way around that. A 60Hz display adds up to 16ms of uncertainty on its own, and a wireless mouse adds a few more. Compare scores on the same machine rather than against a friend on different hardware.
- Can I improve my reaction time?
- Somewhat. Practice, caffeine and being well rested reliably move most people 10 to 30ms, and that is mostly learning the task rather than a faster nervous system. The underlying signal speed barely changes.
- Should I use my average or my best score?
- The median. Your best round is usually a near-miss false start that happened to land after the cue, while the median across five or more rounds is the number that actually repeats.
- Does this test store my results?
- No. Everything runs in your browser, nothing is uploaded, and your scores disappear when you reload the page.
F1 Reaction Time Test
- How do Formula 1 start lights work?
- Five pairs of red lights illuminate from left to right, one per second. Once all five are on, the race director holds them for a random interval between 0.2 and 3 seconds and then extinguishes all of them at once. The race starts at lights out, not at any light coming on.
- What counts as a jump start in F1?
- Moving before the lights go out, or reacting faster than 0.200s after they do. Transponders in the grid slot detect movement, and the FIA treats anything under two tenths as anticipation because human reaction cannot reliably be that quick.
- What is a typical F1 driver reaction time?
- Around 0.2 to 0.3 seconds from lights out. Drivers are not measurably faster than trained members of the public at pure reaction - what they add is a clutch and throttle launch executed perfectly in the same instant.
- Why is the hold before lights out random?
- So the start cannot be counted. If the gap were fixed, every driver would time the launch rather than react to it, and the grid would effectively be a rolling start.
- Why is my F1 score slower than my normal reaction score?
- The anticipation is different. A plain test gives you one colour change; here you watch five lights build and then have to ignore the rhythm you just learned. Most people lose 20 to 50ms to that.
- Does this simulate the car launch too?
- No. This measures the reaction half only. On a real grid the driver also has to release the clutch and feed the throttle without bogging down or spinning up, which is where most places are actually won and lost.
- What is the fastest recorded F1 start reaction?
- Reliable published figures cluster around 0.15 to 0.2 seconds, and anything below 0.2 is penalised, so the practical record is effectively pinned just above two tenths.
- Can I practise this?
- Yes, and it helps. Watching the last light rather than the whole bar, and keeping your finger already resting on the button, are worth more than any amount of raw training.
Fake Tweet Generator
- Is this a real tweet?
- No. It is an image drawn in your browser. Nothing is posted, nothing exists on X, and the image has no link because there is no underlying post.
- Is it legal to use a fake tweet image?
- For parody, teaching, memes, design mockups and slide decks, generally yes. Presenting one as a genuine screenshot of a real person's account is defamation or fraud depending on what it says, and platforms remove manipulated-media posts. Label it.
- Does the image have a watermark?
- No. The PNG is exactly what you see in the preview, at 1200 pixels wide for a crisp result on retina screens and in slides.
- Are my uploads sent to a server?
- No. The avatar is decoded and drawn by your browser on a canvas element. Nothing leaves your machine, and there is nothing to delete afterwards.
- Why does my text look different from a real tweet?
- Real tweets use Chirp, which is not a public web font. This draws in your system UI font, which is close but not identical - fine for mockups, noticeable if you put it beside a genuine screenshot.
- Can I add an image or a quoted tweet?
- Not yet. This covers the common case: avatar, name, badge, text, timestamp and counts. Attached media and quote tweets need a second layout.
- What size is the download?
- 1200 pixels wide, with the height set by how much text you wrote. That is the right size for slides, blog posts and social cards without upscaling.
- Can I use it for a design mockup?
- That is the best use for it. Dropping realistic-looking social proof into a landing page mockup is far quicker than screenshotting a real account, and it avoids shipping someone else's face in your comp.
Fake Instagram Story Generator
- What size is an Instagram story?
- 1080 by 1920 pixels, a 9:16 ratio. This exports at exactly that, so the image uploads without Instagram re-cropping or softening it.
- Does this post to Instagram?
- No. It draws an image. There is no Instagram login, no API call and no account involved - you get a PNG to do whatever you want with.
- Is there a watermark?
- No. The download is exactly the frame you see in the preview, with nothing added.
- Are my photos uploaded anywhere?
- No. The background is decoded and drawn by your browser on a canvas element. It never leaves your device.
- Is it legal to make a fake story?
- For mockups, parody and memes, generally yes. Passing one off as a genuine story from a real person's account is impersonation, and Instagram removes both the post and usually the account. Label it as a mockup.
- Why is the font not Instagram's?
- Instagram's story fonts are proprietary and not available to a web page. This draws in your system UI font, which reads correctly at a glance but will not match pixel for pixel beside a real screenshot.
- Can I use this for a product mockup?
- Yes, and that is the most common use. Showing how your app looks when someone shares it to a story is much faster than building the real share flow just to take a screenshot.
- Where does the caption sit?
- Vertically centred with a soft shadow, which is where most story text ends up and where it stays clear of the top chrome and the bottom reply bar.
Product Hunt alternative
- Can I launch on both?
- Yes, and you should. They do different jobs - Product Hunt is a one-day awareness spike, a verified listing is a page that keeps earning. Nothing in either platform's rules prevents launching on both.
- Does Product Hunt help SEO?
- Mostly indirectly. The spike in traffic and the secondary coverage a good launch generates are worth more than the link itself. What it does not give you is a page that keeps ranking, because your listing stops being crawled meaningfully once the day ends.
- Do I need a hunter?
- Not since self-launching became standard. A well-known hunter still helps with initial visibility, but the practice of paying for one is against the spirit of the platform and increasingly ineffective.
- What counts as a good Product Hunt result?
- Top five on the day is a genuinely good outcome. Below top ten, the traffic difference between rank 12 and rank 30 is small enough that it rarely changes anything.
BetaList alternative
- Is BetaList worth paying for?
- The paid option buys you position in the queue, not better placement or more traffic. If your launch has a hard date, it is worth it. If not, the free queue reaches the same audience.
- How much traffic does a BetaList feature send?
- Substantially less than a good Product Hunt day, and more qualified. Expect hundreds rather than thousands, from people specifically looking for new products to try.
- Can I list a product that already launched?
- The platform is oriented toward pre-launch and early-stage products. A mature product with revenue is outside what the audience is there for.
- Does BetaList verify anything?
- An editor reviews the submission for fit and quality. Nothing you claim about traffic, users or revenue is checked.
Hacker News alternative
- When should I post a Show HN?
- Weekday mornings US Eastern have historically had the most traffic and the most competition. There is no reliable optimum - the variance between identical posts is larger than any timing effect.
- Can I ask people to upvote?
- No. Vote rings are detected and penalised, and the penalty applies to the account, not just the post.
- Does a Show HN help SEO?
- The link itself is nofollowed. The value is in the secondary coverage - blog posts and newsletters that pick up a front-page Show HN and link to you properly.
- What if it gets no traction?
- That is the common outcome and it costs you nothing. Reposting the same product later is tolerated if the product has meaningfully changed.
Peerlist alternative
- Is Peerlist only for developers?
- The audience is predominantly technical. Products for non-technical buyers get noticeably less traction there.
- How does the weekly cycle work?
- Launches accumulate through the week and are ranked within it, which spreads voting out rather than concentrating it in a single day.
- Does Peerlist verify products?
- No. Moderation checks for spam and fit, not for the accuracy of any claim.
- Can I launch on Peerlist and Product Hunt?
- Yes. The audiences overlap less than you would expect, and neither platform restricts it.
Uneed alternative
- Is Uneed better than Product Hunt for a first launch?
- For odds of finishing well, yes. For total traffic, no. If you have nobody to mobilise, a top finish on a smaller board often produces more useful attention than rank 40 on a large one.
- Is the paid option worth it?
- It buys a chosen date rather than better placement. Worth it only if your launch is tied to something else with a fixed date.
- How much traffic does a top finish send?
- Hundreds rather than thousands. The value is more in the listing and the secondary sharing than in the day itself.
- Does Uneed check claims?
- Curation reviews the product for quality and fit. Metrics you state are not verified.
Indie Hackers alternative
- Can I launch a product on Indie Hackers?
- You can post about it, but there is no launch mechanic or leaderboard. Treat it as a place to share what you learned, not as a distribution channel.
- Is the revenue data on Indie Hackers reliable?
- It is self-reported. Most people are honest, and there is no mechanism preventing anyone from inflating a number.
- Does posting there help SEO?
- Not directly. The value is relationships and feedback, both of which lead to links indirectly.
- What performs well there?
- Specific numbers and specific failures. Generic growth advice does not.
Startup Fame alternative
- Are directory links worth chasing?
- A handful of strong, relevant ones are. Two hundred weak ones are not. Check that the page holding your link is indexed and does not carry hundreds of outbound links.
- How is domain rating measured here?
- The figure quoted is a third-party backlink rank on a 0–1000 scale, measured on 29 August 2026. It is not a Google metric and Google does not use it.
- Do paid directory tiers help?
- They buy position on the directory, which affects how many of that directory's visitors see you. They do not change the link's value.
- How many directories should I submit to?
- Ten to twenty good ones, chosen by checking the page that will hold your link. Beyond that the returns fall off sharply.
MicroLaunch alternative
- Is MicroLaunch worth launching on?
- As a warm-up, yes. The traffic is modest, and getting the launch assets right somewhere low-stakes before Product Hunt has real value.
- Do paid boosts work?
- They increase visibility on the board. Whether that converts depends entirely on the product, and the board's total audience caps the upside.
- Can I launch here before Product Hunt?
- Yes, and it is a common sequence. Neither platform requires exclusivity.
- What kind of products do well?
- Small, single-purpose tools with an obvious use. Complex B2B products get less traction.
Fazier alternative
- Are launch badges worth displaying?
- They are social proof, and they create a link back to the platform. What matters is whether the badge stands for something a visitor can check. Most stand for having launched.
- Does a weekly board beat a daily one?
- It spreads voting out, which reduces the timezone lottery. It also produces a lower peak.
- Does Fazier verify claims?
- No. Moderation is for spam and fit.
- Can I use several launch badges?
- Yes, though a wall of badges dilutes each one. Two that mean something beat six that mean you submitted a form.
SaaSHub alternative
- How long does a SaaSHub listing take to matter?
- Months. It is an SEO asset that accrues as the category page ranks, not a traffic event.
- Should I pay for promotion?
- Only once you know the category page sends you traffic organically. Paying to be featured on a page nobody reaches is wasted.
- Do the reviews carry weight?
- With buyers, somewhat. They are unverified, so treat them as social proof rather than evidence.
- Is it a launch platform?
- No. There is no launch day and no leaderboard. Treat it as a permanent listing.
AlternativeTo alternative
- How do I get added to AlternativeTo?
- Anyone can submit a product, including you. What you cannot do is manufacture the votes that decide where it ranks on a page.
- Is it worth the effort?
- The submission takes fifteen minutes and the listing is permanent. That ratio is hard to argue with, even if the traffic takes a long time to arrive.
- Does it help SEO?
- It is a listing on a strong domain and it ranks for switching queries. Treat the referral intent as the main value rather than the link.
- Can I edit my listing?
- Listings are community-editable, so you can improve yours, and so can anyone else.
There's An AI For That alternative
- Is the paid fast-track worth it?
- It shortens the queue. Whether it pays back depends on how well your tool converts directory traffic, which you cannot know until you are listed.
- Does listing there help AI search visibility?
- Indirectly. Being catalogued on a strong domain in your category makes you easier to find and cite, but it is not a substitute for your own pages answering the questions.
- How crowded is it?
- Very. Any generic AI category has thousands of entries, so a narrow, specific positioning matters more here than almost anywhere.
- Do they verify claims?
- Editorial review checks the tool works and fits a category. Metrics are not verified.
Launching Next alternative
- Is a date-ordered directory useful?
- Marginally. The listing exists and is indexed, which is the whole of the value. Nobody browses page forty.
- Should I pay for express submission?
- Only if you have a launch date to hit. It changes when you appear, not how visible you are afterwards.
- How does it compare to stronger directories?
- At a measured backlink rank of 343 it sits below the stronger options in this list. Do the higher-authority ones first.
- Does it verify submissions?
- A light check that the product exists and is not spam.