Comparison
DIY SEO Hub vs Screaming Frog, free against free
Screaming Frog is the desktop crawler most technical SEO people learned on, and its paid version is very good. Most people comparing the two are comparing what they can have for nothing, so that is what the table does: our free scan against Screaming Frog's free version, row by row, including the rows we lose.
Screaming Frog's column comes from its pricing page, read on 17 October 2026. Prices and feature lists change; check theirs before deciding on anything that matters.
Free against free
✓ yes·✓ partly, as the note says·✕ no
| Feature | DIY SEO Hub, free | Screaming Frog, free |
|---|---|---|
| Getting started | ||
| Nothing to install | YesRuns in a browser tab | NoDesktop app for Windows, macOS or Linux |
| Works on a phone, tablet or Chromebook | Yes | No |
| No account, e-mail or licence key needed | Yes | Yes |
| Keeps crawling when you close the laptop | YesThe crawl runs on our servers | NoThe crawl runs on your computer |
| Uses none of your memory or bandwidth | Yes | No |
| E-mails you when the crawl is done | YesOptional: leave an address on the form | No |
| How much it crawls | ||
| Free crawl limit | 5,000 pagesPlus 1,000 images, scripts and stylesheets | 500 URLs |
| To crawl more | Nothing to buyThe limits are the same for everybody | £199 a year per userUnlimited, subject to your machine’s memory |
| Checks | ||
| Broken links, errors and redirects | Yes | Yes |
| Page titles and meta descriptions | YesWith approximate pixel width | Yes |
| Meta robots and X-Robots-Tag directives | Yes | Yes |
| Exact duplicate pages | Yes | Yes |
| Links to other sites checked for errors | YesThe 300 most-linked destinations a scan | Yes |
| Internal link map | Yes | YesSite visualisations |
| Health score out of 100, comparable across sites | YesRanked on the public SEO rank table | No |
| Findings grouped into tasks by shared cause | YesOne task per broken template or dead destination | No |
| Passive security review | YesHeaders, cookies, mixed content, vulnerable libraries | No |
| hreflang audit | YesEvery annotation across the crawl: codes, self-reference, x-default, return links | Yes |
| Generates an XML sitemap | YesFrom the crawl’s indexable pages | Yes |
| HTML and head validation | YesTags outside <head>, repeated tags, the 2 MB Googlebot reads | Yes |
| Soft 404s | YesRead from the title and heading | Yes |
| Pagination (rel next/prev) | YesBroken, unlinked and looping sequences | Yes |
| List mode | YesPaste URLs; their redirects are followed, their links are not | Yes |
| Image sitemap | Yes | Yes |
| Crawl path to a page | YesIn the page inspector | Yes |
| In Screaming Frog’s paid version only | ||
| Save a crawl and open it again later | YesEvery scan stays at its private link | No |
| Scheduled re-crawls | YesWith a free account; e-mails only when something changed | No |
| Compare two crawls | YesWhat changed since the last scan, URL by URL | No |
| AI recommendations | YesA written fix plan from your findings, no API key | NoPaid, with your own OpenAI or Gemini key |
| Search Console data | PartlyImport the Pages export to put the fixes in order | No |
| Backlink metrics | PartlyAhrefs Domain Rating for the site | No |
| Search the source of every page | YesContains, does not contain, or a regular expression | No |
| Custom extraction | YesCSS selector or regular expression, text or attribute, CSV of every page (no XPath) | No |
| JavaScript rendering | YesTicked per scan: compares the served HTML with what a browser builds | No |
| Structured data validation | PartlyJSON-LD checked for Google’s required properties on common rich-result types; microdata and RDFa listed | No |
| Near-duplicate content | YesPages 95% alike, grouped | No |
| Spelling | YesEnglish pages; names and jargon show up too | No |
| Grammar | No | No |
| Crawl settings | PartlyInclude and exclude paths, depth, start folder, all subdomains, sitemap-only; not the user agent or speed | No |
| URL rewriting | YesDrop tracking and named parameters, lower-case paths | No |
| Link Score (internal PageRank) | Yes | No |
| Link positions | YesNavigation, header, footer, sidebar or content, and pages only the template links to | No |
| Accessibility audit | Yesaxe-core, in JavaScript rendering scans | No |
| Rendered page screenshots | PartlyThe first 20 pages of a rendering scan | No |
| Orphan pages from Search Console | YesFrom an imported Pages export | No |
| PageSpeed Insights | YesTop pages of every finished scan, mobile | No |
| Technical support | YesContact form | No |
| Sharing and keeping the data | ||
| Share the results with a link | YesThe person you send it to needs no account and no app | NoSend exports; reopening a crawl needs the app |
| Share anonymously | YesNo name, account or e-mail on the scan | No |
| Public report page for the domain | YesScore and summaries; the detail stays on your private link | No |
| Export to CSV | YesEvery row of every check | Yes |
| The whole crawl as one database file | YesSQLite, with the stored HTML of every page | No |
| Ask an AI assistant about the crawl (MCP) | YesAny MCP client, with the scan’s token | Beside the desktop app |
| Keep a crawl private | YesTick “Keep this scan private”; otherwise a finished scan gets a public summary | YesIt never leaves your computer |
| Where Screaming Frog’s free version wins | ||
| Crawls staging, intranet and localhost sites | NoPublic addresses only, by design | YesIt runs on your network |
| Checks a site the moment you ask | PartlyScans queue when the crawler is busy | Yes |
| Private by default | NoPublic summary unless you tick private | Yes |
Where Screaming Frog is simply better
The paid version. For £199 a year per user you get crawls with no page limit, the Search Console and Analytics integrations through Google's APIs, forms-based login, XPath extraction, grammar checking, full schema.org validation, embeddings and segmentation. If you audit sites for a living, buy it.
It is yours. It runs on your machine, on any site you can reach, including the one behind your VPN, and nothing about the crawl leaves your network unless you send it. Here a scan is public unless you tick private, and only public addresses can be crawled.
Depth of checks. Hundreds of filters against our 37 checks, and the long tail matters on a large site.
Speed. On a desktop with five threads it crawls a site in minutes; a scan here waits its turn and is paced to the site's answers. Rendering JavaScript here is per scan and slower still.
Where this is better
Nothing to install, nothing to sign. Paste a domain and a crawl starts, on any device with a browser. For somebody who needs an answer today about one site, that difference is most of the work.
Ten times the free crawl. 5,000 pages against 500 URLs, and the free scan keeps what Screaming Frog's free version cannot: it is saved at its own link, it can be scheduled, and it can be compared with the next one.
Sharing is a link. Send the private link and the other person sees everything, with no app, no account and no name attached to the scan. Or send the public report.
The data comes out whole. A CSV is the columns somebody chose. Our download is the crawl itself: URLs, links, resources, and the stored bytes of every page, in a file you can query with twenty lines of SQL years later.
It keeps crawling when you close the laptop. The crawl runs on our side, so a big site does not tie up your machine or your connection, and you can be e-mailed when it is done.
Free where Screaming Frog charges. Saved, scheduled and compared crawls, JavaScript rendering with an accessibility audit, source search and extraction, near duplicates, spelling, PageSpeed Insights, Link Score, link positions, URL rewriting and crawl settings are all free here, on every scan.
Things neither version of Screaming Frog does. A passive security review of headers, cookies and library versions; a health score you can compare with every other scanned site; findings grouped into tasks by their shared cause; and a public report page per domain.
Questions people ask
- Is there a free alternative to Screaming Frog?
- This is one, for the common case. Screaming Frog's free version stops at 500 URLs and cannot save, schedule or compare crawls, render JavaScript or extract data; a scan here goes to 5,000 pages and does all of that free. Past 5,000 pages, behind a login, or on a site that is not public, buy the licence: it is inexpensive next to the time it saves.
- What does the Screaming Frog free version include?
- Broken links, errors and redirects; page titles and meta data; meta robots and directives; hreflang; exact duplicate pages; XML sitemap generation; and site visualisations, up to 500 URLs per crawl. Saving crawls, scheduling, crawl comparison, JavaScript rendering, custom extraction, the Search Console and Analytics integrations and technical support need the licence (£199 a year per user).
- Which one should I use?
- If you audit other people's sites for a living, Screaming Frog, and it is not close. If you have one site, want the technical faults found, and would rather not install anything, start here and buy the licence the day you hit a wall.
- Can I get the raw crawl data out of both?
- Screaming Frog exports CSV and Excel, and with a licence can save crawls to reopen in Screaming Frog. Here the crawl is a SQLite database, including the stored bytes of every page, and it downloads with no account at all.
Try the free one first
Read next
- Compared with Sitebulb — the other desktop crawler people ask about.
- Compared with Ahrefs Site Audit — the cloud one inside a bigger subscription.
- What our crawler does — the whole feature list, plainly.