How to check your site for free before buying SEO tools

If you want to check a site for free before paying for SEO software, start with the basics that no subscription can bypass. Can search engines reach the important pages, understand what each page is for, and load the site without obvious friction on mobile? You can answer a surprising amount of that with your browser, Google Search Console, and a few public testing tools.

A free SEO website check is useful when you treat it as diagnosis, not prophecy. It can show blocked pages, missing titles, duplicate intent, poor internal links, weak mobile usability and unstable layouts. It cannot tell you, with certainty, what you will rank for next month, whether an AI overview will mention you, or whether more traffic will become more sales. Those are different questions and need different evidence.

What a free website check can and cannot tell you

A free check can tell you whether the foundations are in place. That includes whether pages return the right status code, whether they are indexable, whether the content appears to match a clear purpose, and whether users can navigate it without confusion. Those are practical checks, and they matter because existing search fundamentals still apply to Google’s AI search features, with no special AI schema or text file required for inclusion in those features, according to Google AI features.

What it cannot do is guarantee outcomes. A tool may score a page 92 out of 100 and still tell you nothing reliable about whether that page deserves to rank for a competitive query. A crawler may report 500 “issues” that are mostly trivial, while missing the real problem, such as three service pages all targeting the same intent.

It also helps to separate evidence types:

• A crawlable page is not the same as an indexed page.
• An indexed page is not the same as a ranking page.
• A ranking page is not the same as a clicked result.
• A click is not the same as an enquiry or sale.
• A citation in an AI answer is not the same as traffic.

That distinction matters because many paid tools blur it. They present technical alerts, visibility charts and estimated traffic on one screen, which can make weak evidence look stronger than it is. Before you buy anything, decide what problem you are trying to solve. If you do not know whether Google can access your key pages, begin there. If access is fine but performance changed suddenly, the next step is investigation, not more dashboards. Our guide on how to investigate a change in search performance is a better next read than a tool comparison.

Check whether search engines can access the pages that matter

Start with the pages that actually matter to your business. For most sites, that means core service pages, key product or category pages, location pages you genuinely use, and important guides or articles.

First, check whether those URLs load normally in a browser. If a key page redirects twice, returns a 404, or loads only after a script-heavy sequence, that is already useful information. You do not need a paid crawler to spot it.

Then check robots rules. Look at the site’s robots.txt file by visiting /robots.txt on the domain. You are looking for broad disallow rules that may affect important sections, such as:

• Disallow: /services/
• Disallow: /blog/
• blocking parameter patterns too aggressively
• staging or development rules left live by mistake

Remember that robots.txt controls crawling, not indexing in every circumstance. A blocked page can still appear as a URL in search if other signals point to it. More importantly, blocking a page in robots.txt is not a substitute for deciding whether the page should exist, be indexed, or be consolidated.

After that, use Google Search Console. In the URL Inspection tool, test individual important pages. You want to know:

• Is the URL on Google?
• Was it crawled successfully?
• Is indexing allowed?
• What canonical did Google select?
• Were there enhancement or mobile usability issues?

Then review the Page indexing report. This is where many site owners either panic or apply blanket fixes. That is usually the wrong approach. Google’s own guidance is to investigate why a URL is excluded, because not every excluded page should be indexed, as explained in the Google Page indexing report.

Common examples:

• “Alternate page with proper canonical” may be fine.
• “Page with redirect” may be expected.
• “Crawled - currently not indexed” needs investigation, especially for important pages.
• “Discovered - currently not indexed” may suggest crawl prioritisation issues, weak internal linking, or low perceived value.
• “Excluded by ‘noindex’ tag” is only a problem if that page should in fact be in search.

This is where a free SEO website check becomes genuinely useful. You can build a short list of affected URLs and sort them into three groups:

1. Expected exclusions
2. Important pages needing action
3. Low-value pages that may be better merged, improved or left excluded

If you want a practical routine for this, see our guide on cutting crawl errors with a practical Google Search Console check.

Review the page itself like a beginner and like Googlebot

Now move from site access to page purpose.

Open an important page and review it twice. First, as a beginner. Second, as Googlebot.

As a beginner, ask simple questions:

• In five seconds, can I tell what this page is about?
• Is the main heading clear?
• Does the title in the browser tab describe the page properly?
• Is the page trying to answer one main need, or several?
• Are the links descriptive, or do they say “click here” and “read more” everywhere?
• If this page is a service page, does it explain the service, the area served, the process, and the next step?

Then think like Googlebot. You are not trying to imitate Google’s systems in full. You are checking whether the page sends clear, consistent signals. The Google SEO Starter Guide remains useful here. It supports straightforward practices such as descriptive titles, helpful content, sensible heading structure and meaningful anchor text.

A few practical checks matter more than long issue lists.

Check titles and headings first. If multiple pages use near-identical title tags and H1s, you may have duplicate intent. For example:

• “Accountancy Services Ipswich”
• “Accountants in Ipswich”
• “Business Accounting Ipswich”

If the content on all three pages is substantially the same and aimed at the same search need, you may not have three opportunities. You may have one topic split into three weaker pages.

Check internal links next. Descriptive links help users and help search engines understand destination pages. “See our payroll service” is stronger than “learn more”. This is not because of magic anchor text formulas, but because it reduces ambiguity.

Check for thin pages. Thin does not mean short. A concise page can be useful. Thin means the page adds little beyond boilerplate, duplicates another page’s purpose, or exists mainly because a template generated it. Typical examples include:

• near-empty location pages
• tag archives with no editorial value
• product variants that should be consolidated
• blog posts written to target a phrase rather than answer a question

Also check whether the visible content depends too heavily on scripts. If key text, links or navigation only appear after complex rendering, free browser checks may reveal that the page is fragile even before you use specialist tools.

A simple way to prioritise is to review ten pages only: your home page, top services, top categories, and your most commercially important articles. If those are unclear, fixing the long tail can wait.

Test speed, stability and usability without overclaiming

Performance testing is useful, but this is where overclaiming is most common.

Use PageSpeed Insights or another public test to look at Core Web Vitals. These metrics describe loading performance, interactivity and visual stability, as documented by web.dev Core Web Vitals. They are helpful because they point to real user friction, such as a page shifting while someone tries to tap a button.

But a score is not a ranking guarantee, and it is certainly not a sales guarantee.

Treat the results as clues:

• A poor Largest Contentful Paint may suggest heavy images, slow server response or render-blocking resources.
• A poor Interaction to Next Paint may suggest script-heavy pages or delayed responsiveness.
• A poor Cumulative Layout Shift score may point to banners, images or embeds loading without reserved space.

You can check a lot for free by hand as well:

• Load the page on your phone using normal mobile data.
• Try to navigate one-handed.
• See whether the menu is clear.
• Check whether text is readable without zooming.
• Watch for pop-ups that cover the screen.
• Try filling in a form.

Usability also includes accessibility basics. The W3C accessibility design tips recommend readable contrast, clear controls and understandable navigation. These are not only compliance or inclusion issues. They are practical quality checks that often reveal why a site feels harder to use than the owner realises.

For UK readers, this is a good point to note that accessibility expectations may differ by sector and legal context. Public sector bodies in the UK face specific website accessibility requirements. Private businesses still benefit from accessible design, but the legal framework is not identical to every EU-wide rule people may have seen discussed elsewhere. If you need legal interpretation, that is separate from an SEO audit.

Do not chase perfect scores for their own sake. A stable, readable, fast-enough site with clear navigation is a better outcome than a technically polished site that still has muddled content and duplicate pages.

Check AI crawler settings without confusing them with Google Search

This area causes more confusion than it should.

Google Search access, Google-Extended, OAI-SearchBot and GPTBot are not the same thing, and you should not treat them as one on-off switch.

For Google, ordinary search crawling is handled by Googlebot. Google’s documentation on AI features says that existing search controls continue to apply, and no special AI text file or special AI schema is required for Google Search AI features Google AI features. In other words, if your pages are accessible to Googlebot and otherwise eligible for search, there is no separate “turn on AI results” file to add.

Google-Extended is a separate control related to certain uses of content by Google. It is not the same as allowing or blocking Googlebot for Search.

For OpenAI, the distinction is also important. According to the OpenAI crawler documentation, OAI-SearchBot is used for search and retrieval, while GPTBot relates to model training. Those are separate choices. Blocking GPTBot does not automatically block OAI-SearchBot, and allowing OAI-SearchBot does not mean you have opted into everything else.

Before changing anything, check your robots.txt for entries such as:

• User-agent: GPTBot
• User-agent: OAI-SearchBot
• User-agent: Googlebot
• User-agent: Google-Extended

Then ask a simple policy question. What are you trying to allow or prevent?

• If you want normal Google Search visibility, do not block Googlebot from important content.
• If you want to control Google-Extended separately, do that deliberately.
• If you want OpenAI search systems to access public pages, review whether OAI-SearchBot is allowed.
• If you want to restrict training access while allowing search retrieval, that is a separate configuration from search access.

This is one of the few areas where a free check can prevent an expensive misunderstanding. We have seen site owners think they blocked “AI” when they actually blocked search, and others assume they are available to AI search because their pages are public, without checking crawler access at all. If this topic is your main concern, our AI search and visibility guide is the better next step.

Turn the findings into a sensible next action

By this point, you should have a short list of findings, not a giant spreadsheet.

The useful question now is not “which tool should we buy?” but “what type of problem do we actually have?”

Usually it is one of four things.

First, a technical access problem. Important pages are blocked, redirected badly, canonicalised away, noindexed by mistake, or simply not being indexed. If so, fix those issues first. Buying a rank tracker before resolving access problems is backwards.

Second, a content and intent problem. Pages exist, but they overlap, say very little, or fail to answer the need implied by the query. In that case, rewrite, merge or retire pages before you buy more reporting.

Third, a measurement problem. The site may be functioning, but you cannot tell which pages changed, which queries fell, or whether the drop is brand, non-brand, local, mobile, or page-specific. Then your next step is better use of Google Search Console and cleaner reporting, not necessarily a full SEO suite. Our piece on Search Console evidence and its limitations can help set expectations.

Fourth, no urgent problem at all. Many sites are simply early. New pages can take time to be crawled, assessed and compared with alternatives. If the site is accessible, the pages are clear, and the measurement is stable, patience may be more sensible than a new subscription.

A practical next action looks like this:

• list ten priority URLs
• check indexing status in Google Search Console
• review each page title, H1 and internal links
• note duplicate intent and thin content
• test mobile usability and layout stability
• review robots.txt and relevant crawler permissions
• decide whether the next task is technical, editorial, measurement, or wait-and-watch

If you want a repeatable routine after this first pass, our monthly SEO check a small business can actually keep up is designed for exactly that. It keeps the focus on evidence rather than tool noise.

The point of a free check is not to avoid ever paying for tools. It is to avoid buying them before you know what question you need them to answer. That is a better habit for beginners, marketers and business owners alike, and it is exactly the sort of evidence-led approach we publish on The SEO Blog, an editorial publication by Fleeta Limited in Ipswich. If you later need practical implementation, that is where Suffolk Web, our sister agency under the same owner, may be relevant. But before any quote, the first job is still the same, check what the site is actually doing, and what the evidence really shows.