Getting started

Crawlix crawls your website, finds the technical and on-page problems that stop search engines and AI assistants understanding your pages, writes the fix, applies it where you allow, and then re-measures the page afterwards to check whether the change actually helped — reverting it if it made things worse.

That last part is the unusual one. Most SEO tools stop at the list of problems and leave the work, and the question of whether it worked, to you.

• Crawls your site and detects 91 distinct issue types from your real pages
• Writes a specific fix for a specific page, using AI
• Applies that fix through your CMS, at the edge, via the snippet, or as a GitHub pull request your team reviews
• Lets you undo any change it made
• Records what it expected to happen, then measures what did
• Tracks where you rank for keywords you choose — this needs a rank-data provider on your plan

Being clear about this up front saves you planning around something that is not coming.

• Guarantee a ranking. Nobody can. Rankings are Google's decision and depend on competitors nobody controls
• Make an AI assistant cite you. Crawlix improves what those systems read about you; the decision is theirs
• Acquire backlinks or create brand demand. Where one of those is what is holding a keyword back, Crawlix says so and says plainly that it cannot move it
• Manage your Google Business Profile, reviews, or the map pack
• Edit the body content of your pages through most CMSes. See the CMS table below — this one surprises people

Have these ready:

• The exact domain you want scanned, in the form you consider canonical. https://www.example.com and https://example.com are different projects
• Admin access to your CMS, if you want Crawlix to apply fixes for you
• Access to your DNS, or the ability to upload a file to your web root, for domain verification
• Google Search Console access, if you want indexation and query data

Sign up with an email and password, or a social login. You will land on the onboarding flow, which walks through the next three steps.

Enter your domain and choose your platform.

Verification proves you control the site. Two methods: a DNS TXT record, or a file uploaded to your web root.

Verification is required for Autopilot and recommended for everything else. Until it passes, Crawlix will scan and advise but will not apply anything automatically.

This is how Crawlix writes fixes back to your site. Six integrations are supported, and they are not equally capable:

Read that last column carefully. Every CMS integration writes titles, meta descriptions, canonicals, schema and alt text. None of them can edit the body of your pages. Only the GitHub pull-request route can, because it patches your page's actual source file — and a person reviews the diff before it merges.

If a content change cannot be delivered to your origin, Crawlix refuses to publish it rather than putting it somewhere search engines will never see.

The snippet does two things: it collects real Core Web Vitals from your actual visitors, and it can deliver certain fixes.

You can skip this and add it later. Without it you still get lab performance data; you just do not get field data from real visits.

Press scan. You will see live progress as pages are crawled. How long it takes is proportional to the size of your site.

When it finishes you get a score, an issue list grouped by severity, and a set of proposed fixes. Reading your first audit is next.