What Free SEO Audit Tools Actually Show You (and What They Miss)

Most free audit tools hand you a 0-100 score and call it a day. Here's what that score is actually built from, where free tools quietly stop being useful, and a worked audit of a real site.

A "95/100 SEO score" tells you almost nothing on its own. It's usually a weighted rollup of pass/fail checks title present, meta description present, no broken links, images have alt text and none of those checks know whether your content is actually good, or whether the page you're auditing is even the one Google has chosen to index for that URL cluster. The score is a compliance report, not a ranking prediction.

Here's what's actually behind that number, where free tools stop being enough, and a walkthrough of auditing a real site.

What a free audit tool is actually measuring

Most free tools including our own pull from one of two places: a live crawl of your site (fetching pages and parsing the HTML directly) or Google's own APIs (PageSpeed Insights/Lighthouse for performance, and if you've connected it, Search Console for indexation and query data). The distinction matters because it determines what the tool can and can't see.

A crawl-based check can tell you a page is missing a canonical tag. It generally cannot tell you whether that page is actually indexed for that you need Search Console's URL Inspection data, which reflects what Google's index actually has, not what your HTML currently says. A page can have perfect on-page tags and still sit in "Crawled – currently not indexed" if Google's quality systems decided it wasn't worth keeping. No crawler-based free tool will surface that; only Search Console will.

Lighthouse-based Core Web Vitals numbers are lab data a simulated run under fixed network/CPU throttling not the real-user field data Google actually uses for the ranking signal (which comes from the Chrome UX Report, aggregated real visitor data). A page can score 95 in Lighthouse and still fail the actual CWV assessment in Search Console's "Core Web Vitals" report if your real visitors are mostly on slower connections or older phones than the lab simulation assumes. If you only check Lighthouse, you're auditing a simulation, not your actual users.

Where free tools hit a wall

Free crawlers cap out somewhere between 100–500 URLs on most tools before requiring payment or a signup wall. Below that ceiling, a crawl-based audit genuinely covers a small site end to end. Above it, you're only ever seeing a sample, and the sample a free crawler chooses (usually breadth-first from the homepage) may never reach your deepest paginated or faceted-navigation URLs which, on an e-commerce site, is exactly where duplicate-content and crawl-budget problems tend to live.

The other wall is log files. A crawl tells you what could be found by following links. Server log files tell you what Googlebot actually requested, how often, and which pages it's ignoring despite being linked and crawlable. That gap pages you think are being crawled regularly versus pages Googlebot has quietly deprioritized only shows up in log data, and free tools don't have access to your server logs by definition. If you're debugging why a large site isn't getting new pages indexed promptly, this is usually the actual bottleneck, and no amount of on-page tinkering will show up in a free audit's report.

The priority order that actually matters

Fixing issues in the wrong order wastes time. Indexation problems come first because everything else is irrelevant if the page isn't eligible to rank at all:

Indexation and crawl blocks. A noindex tag or a robots.txt disallow rule left over from staging silently removes a page from consideration check this before anything else, because a perfect page behind a noindex ranks for nothing.
Duplicate or near-duplicate title/meta tags. These cause Google to guess which of your pages should rank for a query, which is the mechanism behind keyword cannibalization two of your own pages splitting authority instead of one page consolidating it.
Broken internal links and redirect chains. Each hop in a redirect chain (A → B → C) dilutes the link equity passed through and wastes crawl budget on sites with limited crawl allowance.
Core Web Vitals failures on real-user data, not just lab scores check the Search Console CWV report specifically, not just a Lighthouse run.
Structured data. Useful for SERP appearance (rich results, FAQ snippets) but it doesn't affect whether the page ranks in the first place, which is why it belongs last, not first, despite showing up prominently in most audit tool dashboards.

A worked audit

Take a hypothetical mid-size site a SaaS company with around 800 published pages, mixing marketing pages, blog posts, and a help center.

Indexation first: pull the Search Console "Pages" report, sort by "Not indexed," and read the reason column. If a chunk of blog posts show "Crawled – currently not indexed," that's a content-quality signal, not a technical bug no crawler-based free tool will explain why, only flag that the pages exist.
Duplicate tags: export all title tags and meta descriptions (most CMSs allow a bulk export, or a crawl tool will list them), and sort to find exact or near-duplicates. On an 800-page site, this is almost always where a templated content system (a fixed blog post skeleton, a boilerplate product page generator) has quietly produced dozens of near-identical tags.
Redirect chains: check for any URL that 301s more than once before landing common after a site migration or a URL structure change that wasn't fully cleaned up.
Core Web Vitals: compare the Search Console field-data CWV report against a fresh Lighthouse run on the same URL. A large gap between the two usually means your real traffic skews toward slower devices or connections than the lab test assumes worth knowing before you spend engineering time chasing a lab score that isn't representative.
Log files (if you have server access): pull 30 days of Googlebot requests and check crawl frequency against your most important commercial pages. If Googlebot is spending most of its budget on parameterized URLs or old paginated archives instead of your priority pages, that's a crawl-budget problem no free front-end audit tool will ever surface.

Two things worth correcting

"Free tools use worse data than paid tools." Not always true for the performance layer specifically most free tools, including PageSpeed Insights itself, pull from the same Lighthouse and Chrome UX Report data that paid tools use under the hood. What paid tools add is usually scale (crawling beyond the free URL cap), log file analysis, and historical trend tracking not a fundamentally different data source for Core Web Vitals.

"A perfect audit score means the page will rank." A 100/100 technical score means the page has no structural reasons not to rank. It says nothing about whether the content itself is better than what's currently in position one through ten for that query that comparison isn't something a technical crawl can evaluate, because it requires reading the competing pages, not just your own HTML.