You paste a URL. You click Scan. Forty-five seconds later you've got a report.
What's happening in those 45 seconds? Honestly — a lot. And I want to walk you through it, because I think you should know what a tool is doing to your site before you trust its findings.
No black box. No "proprietary AI algorithm." Just the actual steps, in order.
Step 1: We fetch your page like Googlebot does
The scanner starts by requesting your homepage, but not the way your browser does. We identify as a bot, the same way Google's crawler does. Your server decides whether to let us in or block us.
This alone tells us something. If your site returns a 429 Too Many Requests to a single scan, Google is probably hitting the same wall when it tries to crawl you. If your site returns a 200 OK in 300 milliseconds, you're in good shape. If it takes 4 seconds to respond, we note that — because Google does too.
Quick side note: if you've got a bot-blocker that refuses our scan, it's probably also making life harder for Google. That's a whole separate finding we'll flag for you.
Step 2: We read your HTML the way a robot reads it
A browser renders your page into something beautiful. A search engine reads the raw HTML and extracts meaning from it.
So that's what we do. We pull the raw HTML and extract:
<title>tag<meta name="description">- All
<h1>through<h6>headings - Every
<a href="...">link (internal and external) - Every
<img>tag and itsaltattribute <link rel="canonical">(if present)- Open Graph and Twitter Card metadata
- Schema.org structured data
- The
<html lang="...">attribute - Robots directives (meta and HTTP header)
This is the same information Google pulls. If anything here is missing or broken, Google misses it too.
I'll tell you the most common thing we find in a second. Spoiler: it's boring. But it's costing people real traffic.
Step 3: We check what you told robots to do
Before we crawl any further, we respect your rules. We look at:
- robots.txt — the file at
yoursite.com/robots.txtthat tells crawlers where they can and can't go - Meta robots tags — page-level directives that can say "don't index this"
- X-Robots-Tag HTTP header — the same thing but sent in the server response
Here's where we catch a lot of accidental self-sabotage. I've seen sites with Disallow: / in their robots.txt — that tells every search engine "don't crawl anything." The site was invisible on Google for months and nobody knew why. One line of text. Gone.
Step 4: We follow your internal links
From your homepage, we follow every internal link and do the same analysis on each page. We cap it — we're not trying to crawl 10,000 pages in 45 seconds — but we get a representative sample.
For each linked page we check:
- Does it return a 200? (or is it a 404, a redirect chain, a timeout?)
- Does it have its own unique title and meta description?
- Are its images alt-tagged?
- Does it duplicate another page's content?
- How fast does it load?
The part people don't realize: a single broken link inside your site isn't just bad UX. Google uses internal links to understand which pages matter. A page with no incoming internal links is an orphan — Google barely knows it exists. A page with 40 incoming internal links is, in Google's eyes, the most important page on your site. Whether you meant it to be or not. (I went deeper on why broken links cost more than you'd think over here.)
Step 5: We check the performance stuff
Page speed isn't a single number. It's a bundle of measurements — Largest Contentful Paint, Cumulative Layout Shift, Interaction to Next Paint. Google calls these Core Web Vitals, and they're a real ranking signal now.
We measure:
- Time to first byte (how quickly your server responds)
- Total page weight (how many megabytes a visitor has to download)
- Image sizes (are you serving a 4MB hero image when 200KB would look identical?)
- Render-blocking resources (scripts that stop the page from showing until they finish loading)
We don't run a full Lighthouse audit in 45 seconds — that would take several minutes per page. But we catch the obvious stuff, and the obvious stuff is what's hurting most small sites.
Step 6: We check accessibility too
This might surprise you. What does accessibility have to do with SEO?
More than you'd think. Google has confirmed that accessibility practices overlap heavily with good SEO: clear heading hierarchy, alt text on images, descriptive link text, proper form labels, readable contrast. A site that works for a screen reader also tends to be a site search engines understand well.
Plus — the legal risk of an inaccessible site in 2026 is not zero. ADA lawsuits are up. Better to find out now.
Step 7: We assemble the report
Everything above gets scored, categorized by severity, and written up in English. Not "H1 hierarchy non-compliance detected." Actual sentences like: "Your About page has two <h1> tags. There should only be one. Here's which one to change."
Free version gives you the list of issues and their severity. $8.99 remediation guide gives you the exact fix for each one — which plugin setting to change, which line of code to paste, which WordPress checkbox to tick.
The boring truth most scans reveal
You want to know the single most common finding across thousands of scans? Missing or duplicate page titles. Really. That's it. The thing every SEO guide has told people to fix since 2005. (Here's the five-second fix I keep begging people to make.)
Second place: missing image alt text.
Third: pages blocked from Google by accident.
None of these are glamorous. None of them require you to hire an agency. They're the kind of thing you can fix on a Sunday afternoon with a cup of coffee and the right list.
Which is the whole point of the scanner. Tell you what's actually broken, in plain English, with the fix right next to the diagnosis. No homework. No upsell.
Run one on your site at abbyseo.com. Free, no signup, 45 seconds. If the report's clean, that's great news. If it's not, now you know — and the fix is waiting for you.