Find why nobody arrives. Fix it yourself in one paste.
The SEO look-over, without hiring an agency.
Your builder made you a site. It did not make you findable. Paste your web address and we fetch your pages the way Google does, then tell you which ones arrive empty, which search engines cannot list, which are found but never clicked, and whether an AI assistant can quote you at all. Every finding ends in a prompt you can paste into whatever you built the site with.
One site, one flat price. Eighty-six checks across eleven segments, scored on what your pages actually send. Every claim names the exact page and line it came from, and is re-checked against that page before you see it.
The same page, twice.
Your builder makes a page that assembles itself in the browser. That works for a visitor who is already there. It is not what arrives at anything that has to decide whether to send you one.
What you see
Your browser, after it runs your JavaScript
What a search engine gets
The HTML your server actually sent, first pass
<body> <div id="root"></div> <script src="/assets/index.js"></script> </body>
No heading. No sentence. Nothing to index, nothing to quote.
We read the second one, because that is what every one of these reads before it decides anything about you:
- Bing
- ChatGPT
- Claude
- Perplexity
- the preview in a WhatsApp message
If your site looks perfect to you and empty to us, that gap is the finding, and it is the most expensive one there is, because it is invisible from the only place you ever look at your own site.
This is what a finding looks like.
On the left, what your page actually sent us. On the right, what it costs you, and the prompt that fixes it.
harpergoods.com/contact
6 <head>7 <meta charset="utf-8" />8 <title>Contact Page</title>9 </head>10 <body>
Your contact page is called “Contact Page” in search results
/contact:8
The title is the blue line a person reads before deciding whether to click. Yours says Contact Page, which tells them nothing about who you are or what they can ask you. There is no description either, so the search engine writes its own summary by pulling a sentence off the page.
On my /contact page, the title is still "Contact Page" and there is no description. Change the title to "Contact Harper Goods: order help, returns and wholesale" and add a description of about 25 words saying who we are and what someone can get in touch about. Both need to be in the page the server sends, not added by JavaScript afterwards.
What we score
Eleven segments, each scored as the share of its checks your site passed. No weighting, no invented number: 72 out of 100 means you passed 13 of the 18 checks in that segment, and the report names all five you did not.
| Technical SEO | Whether search engines can reach, read and index your pages at all |
| On-Page SEO | What each page says about itself: titles, summaries, headings, pictures |
| Generative Engine Optimization | Whether an AI writing an answer has anything of yours to use |
| AI Overviews | Whether you turn up inside the answer box above the results |
| Answer Engine Optimization | Whether your pages answer questions directly enough to be quoted |
| LLM Optimization | Whether assistants are allowed to read you, and find something worth reading |
| E-E-A-T | The signals that say a real business with real people stands behind this |
| Programmatic SEO | Whether you have pages built from a template to cover many searches |
| Off-Page SEO | What your own site says about your presence everywhere else |
Two of those we can only see one side of, and the report says so where it scores them. Nothing in a page your server sends reveals who links to you, so Off-Page is scored on the presence your own site declares, and E-E-A-T on the trust signals your pages carry rather than on whether anyone finds you authoritative.
Questions people ask before paying
What do I actually get for $49?
One audit of one website. Eighty-six checks across eleven segments, a score that is a plain count of what passed, every failed check named with what we saw, and a written explanation plus a copy-paste fix prompt for the ones worth acting on. The report stays at its link.
Do I need to understand code?
No. Every finding is written for the person who owns the business, not the person who wrote the site, and the fix is a paragraph you paste back into Lovable, Bolt, Replit or whatever built it. If a developer looks after the site, there is a brief you can send them instead.
How long does an audit take?
A few minutes. You can close the tab: the report arrives by email if you asked for it, and sits on your account either way.
Will this get me to the top of Google?
Nobody can promise that, and we do not print a traffic or ranking forecast anywhere, because no such number can be derived from what a page sends. What we can tell you is exactly what your pages send a search engine, and what to change about it.
What this audit will not do.
It reads what your site sends, so it can only tell you things your own pages prove. It will not count your backlinks, check who mentions you, or promise you a position in the results. Anyone selling you that is guessing.
It also reads your pages without running their JavaScript, because that is what a search engine does on its first visit. If your site looks perfect to you and empty to us, that gap is the finding.
What it does instead is name the things you control, on the pages you own, with the exact change to make. If you would rather we read the repository behind the site, you can paste that instead.