How I Got DIY SEO Hub Close to Parity With Screaming Frog's £199-a-Year Licence

What the £199 buys

Screaming Frog’s SEO Spider is free up to 500 URLs a crawl, and its pricing page (read on 17 October 2026) lists one licence at £199 a year per user, about US$270 at the time of writing, with volume discounts from five seats. Officially the only difference between free and paid is the crawl limit. In practice the licence is also what unlocks the things people actually pay for: saving a crawl and opening it later, scheduling, comparing two crawls, JavaScript rendering, custom extraction, the Google integrations and support. Those are the rows the free version greys out, and those are the rows I set out to match.

I want to be precise about the claim, because the comparison is easy to inflate. Screaming Frog’s paid version is a better crawler than mine for anyone who audits other people’s sites for a living, and I say so on the comparison page. The claim is narrower: for someone with one site, who wants the technical faults found and does not want to install anything, DIY SEO Hub now covers most of the paid column for nothing.

The paid-only column, row by row

These are the features Screaming Frog’s own pricing page reserves for the licence, scored against DIY SEO Hub as of today. An amber tick means partly, with the limitation stated; the live free-against-free comparison on the site is the maintained version of this table and includes the rows where the free Screaming Frog wins outright.

How the rows got built

Nothing here is clever on its own; the work was deciding what “done” meant for each row, then building exactly that. A few of them are worth describing because they shaped the rest.

Saving, scheduling and comparing were nearly free, because every scan was already its own SQLite database at a private link. Scheduling is a cron row against an account; comparison is a query that joins two scans on URL and reports what changed in status, redirect, title, description and indexability. Screaming Frog has to add these to a desktop app; here the architecture handed them over.

JavaScript rendering is a per-scan tick, because rendering is slow and most sites do not need it. A rendering scan fetches the served HTML and what a headless browser builds, then diffs them, which is the question people have about JavaScript sites: is what Google sees what the user sees? The same browser session runs the axe-core accessibility audit and keeps screenshots of the first twenty pages.

Source search and extraction run over the stored bytes of every page. Because the crawl keeps the page bodies, a search for a string, a regular expression, or a CSS selector never touches the site again. Extraction gives you a CSV with one row per page. I did not build XPath, and the table says so.

The Google integrations are where I stopped short on purpose. Screaming Frog connects to Search Console and Analytics through Google’s APIs, which means OAuth, scopes and a verification process. DIY SEO Hub takes the Pages export from Search Console as a file, and uses it for two things: putting fixes in the order of the pages Google actually shows, and finding orphan pages that Google knows about but the crawl could not reach. It is less, and it needs no permission from anyone.

Link Score and link positions came from the link table that was already there. Internal PageRank over the stored link graph is a few lines; classifying each link as navigation, header, footer, sidebar or content needed the rendered DOM and became the first feature that paid for the rendering work twice.

AI recommendations write a fix plan from the findings. Screaming Frog’s equivalent needs your own OpenAI or Gemini key; here the plan is included, and it is advice to verify rather than a to-do list to execute, which the page says in as many words.

Where it still loses

The comparison page has a section headed “Where Screaming Frog is simply better”, and it is not decorative. The licence gives you crawls with no page limit where mine stops at 5,000; forms-based login for sites behind authentication; XPath; grammar; full schema.org validation; embeddings and segmentation; and hundreds of filters against my few dozen checks, which matters on a large site. It runs on your machine, so it can crawl staging, an intranet or localhost, and nothing leaves your network. It is also faster: five threads on a desktop crawl a site in minutes, while a scan here waits its turn and is paced to the site’s answers.

And a scan on DIY SEO Hub is public unless you tick private. That is a deliberate trade for the public report page per domain and the health score you can compare across sites, but if private by default is what you need, the desktop app is the right answer and the table says so.

Why it is free

DIY SEO Hub is a product of 8Examples, the same company that builds managed AI workers and websites for Calgary small businesses. A free crawler that finds the faults on a small business site, and shows the owner exactly which page to fix, is the most honest way I have found to introduce ourselves. The limits are the same for everybody, there is no paid tier to upsell into, and the crawl downloads as a database you can keep.

If you have a site with old links, uneven titles, or pages that are hard to reach, run it through the Site Spider, then read the comparison and decide for yourself whether the licence is still worth £199 for your situation. For a lot of sites it is not. For an agency it still is, and it is not close.