Best technical SEO tools and audit software compared, ranked on the ceiling each vendor publishes

This page is about the software. If you are buying the audit as a service, the deliverable is the thing to compare rather than the tool, and that is answered separately. If what you want is the order to run the checks in, that is a checklist rather than a purchase.

The criterion on this page, applied to everybody including us. A technical tool exists to read your site at scale, so rank on the ceiling the vendor publishes and on whether JavaScript rendering is included or charged for. Both are the vendor's own numbers, both are checkable today, and neither moves with the country you read them from. Everything else in this category is a feature grid.

What each vendor publishes, read 21 October 2026

The contradiction on the Screaming Frog page, reported rather than resolved

On 21 October 2026 that page says in prose that "the crawl configuration, saving of crawls, and advanced features such as JavaScript rendering" are restricted in the free version, and that a licence "allows you to save crawls, and opens up the spider's configuration options and advanced features". Its own feature table on the same page lists JavaScript rendering, crawl configuration, saved and reopened crawls, scheduling, structured data validation and the Analytics, Search Console and PageSpeed integrations on the free side, with the crawl limit as the only stated difference between free and paid.

Both readings are from the same page on the same day. We are reporting both, because picking the one that suited an argument is how a comparison stops being one. Install the free version and check the feature you care about before planning around either reading.

So the ranking, on the stated criterion

Sitebulb first. It is the only vendor here that publishes a per-audit URL ceiling for each desktop tier and states that JavaScript crawling is included at no extra cost. That is both halves of the criterion, in numbers, on the page a buyer reads.

Screaming Frog second. Its free ceiling is published and exact, which is more than most vendors manage, and the paid ceiling is honestly described as a function of your own machine. It loses first place only because its own page cannot agree with itself about what the licence unlocks.

Oncrawl third, and first for anybody whose real problem is log files. It publishes scale rather than entitlement, which is a weaker form of disclosure, and its log analysis is a genuine capability the two desktop tools do not have.

Lumar and seoClarity are not ranked on this criterion. seoClarity publishes "unlimited" and no number, which is a claim rather than a limit; Lumar publishes neither. Both are built for sites past the point where a desktop crawl is practical, and if your site is that size the demo call is the route whatever this page says.

Search Console sits outside the ranking and should still be open before any of them. It reports which of your pages Google actually indexed and why it declined the others, and every crawler above is inferring what it tells you directly.

The eight checks a technical audit has to produce, whatever tool runs it

Crawl by template rather than by URL. Render comparison between raw HTML and the rendered DOM. Index coverage from Search Console, which is first party and cannot be substituted. Log file reading, which is the check that separates an audit from a checklist. Core Web Vitals from field data rather than lab data. Internal linking with orphan pages listed and the denominator stated. Structured data validation with errors separated from warnings. Duplication and URL parameters, with a canonical decision per pattern.

A tool that produces seven of those and a vendor score is still a good buy. A report that produces a score and none of them is a score.

The one check none of these tools can run, and who sells it

Everything above reads your site, which is a thing you own and a crawler can fetch. The other place your buyer now asks is an assistant, and that answer is composed fresh for each person. ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews name some businesses and not others, and there is no index of it to crawl, because nothing was stored. It is outside the scope of a technical audit by definition, since it is not a property of your site at all.

AI Knows Us sells that one measurement and nothing else, and we are deliberately not in the ranking above, because a buyer who came here for a crawler should buy Sitebulb or Screaming Frog. We are not a technical SEO tool. From the rendered text of our pricing page, read 21 October 2026: "One credit is one buyer question, asked on one AI app, once", "Most checks use one credit", "A Claude check uses 3, because Claude costs us more to run". From our llms.txt the same day: SEO Start at $149 a month watches 0 buyer questions, 0 credits and 0 engines, so it is a demand-harvest tier rather than an engine watch, and Growth at $299 with 100 buyer questions and 1,200 scheduled answers is the cheapest plan that reads all five surfaces. A managed option is published in rupees only at Rs 88,499 to Rs 2,35,999 a month with a three month minimum.

Where these vendors beat us on our own criterion. Three of them now report on AI crawlers or AI citations from the server side, which is a different and complementary measurement to ours: Oncrawl's AI Search Lens "uses real crawl and log data to show which parts of your site LLMs actually cite", Lumar publishes a GEO Toolkit, and seoClarity publishes an AI search module. Server-side bot data is something we do not read at all, and if that is the question, a log platform answers it and we do not. Ahrefs Brand Radar, read the same day, publishes "1 check per platform, location and update", "$0.020/check" and "Claude consumes 8 checks per update", and covers AI Mode and Copilot, which are two surfaces we do not read.

For the one free check that overlaps with all of this and is misread more than any other number in technical SEO: Lighthouse SEO, the ten checks behind the score and why one of them outweighs the rest. On 21 October 2026 we read 2,411 page bodies on this site against 2,325 live pages and 971 of them, 41.8 percent, carried no inbound link from any body we could read, which is an upper bound rather than a count. We measured that target specifically: no other page on this site links to that one.

Questions about technical SEO tools

Do I need a desktop crawler and a cloud platform?

Not until the crawl stops finishing on your own machine. Desktop handles staging sites and password-protected builds, which cloud crawlers find awkward, so most teams past a certain size end up with both for different jobs.

Is a free crawl enough?

For a site under five hundred URLs, the published free ceiling is the whole site, so yes. Past that, you are sampling, and a sampled crawl hides exactly the template that is broken.

Does rendering matter if my site is server rendered?

Check rather than assume. Most modern stacks render some components client side, and the render comparison takes one run.

Which number should decide the tier I buy?

Your URL count, taken from your sitemap and then doubled, because the crawl finds parameters and pagination the sitemap does not list.