How to check the SEO of a website
To check the SEO of a website, look at seven things: the title tag and meta description, the heading structure, whether Google has the pages indexed, the robots.txt file, mobile rendering, load speed, and structured data. You can check each one by hand in a browser, or run a free scan that checks all of them at once.
What checking the SEO actually covers
Three questions decide most of it. Can Google find and index the page? Does the page say clearly what it is about? Does the page work, meaning it loads fast and reads fine on a phone? Every web SEO checker ever built automates some version of that list. The good news is you can check website SEO by hand with nothing but a browser and a Google search box. Here is the list, in the order that matters.
1. Title tag and meta description
Right-click the page, choose "View page source", and find the <title> tag near the top. It should be unique to that page, say what the page is about in words a customer would type, and put the topic at the front. A homepage titled "Home" tells Google nothing, and Google is what sends you customers. The meta description sits nearby: a plain one-or-two-sentence summary. Google rewrites descriptions often, but a good one still earns the click when it survives. Do not obsess over exact character counts; Google truncates titles by available pixel width, not a fixed number. If yours are broken, start with fixing the title tag and the meta description.
2. Headings
In that same source view, search for <h1. There should be one main heading that matches what the page is about, then <h2> sections breaking the content up in a sensible order. Machines lean on that structure to understand the page. A wall of styled <div> text with no real headings reads fine to you and poorly to a crawler. Details in our heading structure fix.
3. Indexation
Search site:yourdomain.com on Google. That shows roughly which of your pages Google has in its index. If the pages that make you money are missing, nothing else on this list matters yet, because a page that is not indexed cannot rank at all. The usual culprits are a noindex tag left over from staging or a canonical pointing somewhere wrong. Google Search Console, which is free, tells you exactly why a page is out. See fixing accidental noindex.
4. robots.txt
Open yourdomain.com/robots.txt in your browser. You are looking for two things. First, a catastrophic Disallow: / under User-agent: *, which tells every crawler to stay out of the whole site. Second, blocks on specific AI crawlers like OAI-SearchBot or PerplexityBot, which quietly remove you from AI answers while your Google rankings look untouched. Our robots.txt fix guide walks through the rules line by line.
5. Mobile
Open the site on your actual phone, or press F12 in Chrome and toggle the device toolbar. Google predominantly indexes the mobile version of your site, so what you see on that small screen is closer to what Google sees than your desktop view. Check that text is readable without zooming, buttons are tappable, and nothing overflows the screen. A site that frustrates a phone user loses that customer twice: once in the ranking, once in the visit.
6. Speed
Paste your URL into PageSpeed Insights at pagespeed.web.dev. It is free, made by Google, and shows both lab measurements and real-user data where available. Pay attention to the largest contentful paint number: that is how long a visitor stares at a blank or half-loaded page before your content appears. Slow pages lose visitors before the page ever gets a chance to sell. Common fixes are in our LCP guide.
7. Structured data
Back in view-source, search for application/ld+json. That is your structured data, the machine-readable block that tells Google and the AI engines exactly what your business is, what the page is, and how it all connects. If it is missing, or if pasting it into validator.schema.org throws errors, you are making every machine guess. Start with the structured data fix.
What manual checking misses
The checklist above covers the classics, and it is worth knowing how to run it. But it has two problems. It takes an hour or more per site if you are honest about it, and it stops where traditional SEO stops. It will not tell you whether your content survives without JavaScript, whether each AI engine's crawler can reach you, or whether your pages contain answers an engine can actually lift. That is a growing share of how customers find businesses now, and it fails silently. AuditLamp's free scan at auditlamp.com/app runs 116 documented checks in about a minute: everything on this page, plus the AI visibility layer that manual checking misses, ranked by impact in plain language. If you want the deeper tiers after that, they are on the pricing page. For the AI side specifically, our guide on checking your AI search visibility is the companion to this one.
Frequently asked questions
What is the fastest way to check the SEO of a website?
Run an automated scan. A manual check of titles, headings, indexation, robots.txt, mobile, speed, and schema takes an hour or more per site. AuditLamp's free scan runs 116 documented checks in about a minute and ranks the problems by impact.
Can I check website SEO for free?
Yes. Everything in the manual checklist uses free tools: your browser's view-source, a site: search on Google, PageSpeed Insights, and the schema validator. AuditLamp's scan is also free to run, so the only real cost of checking is your time.
How do I check if Google has indexed my site?
Search site:yourdomain.com on Google. If pages you care about are missing, Google Search Console tells you why, and it is free. A page that is not indexed cannot rank at all, so this is the first thing to rule out.
How often should I check a website's SEO?
After every meaningful change to the site, and on a regular schedule in between. Redesigns, migrations, and plugin updates silently break titles, indexation, and schema. Owners usually find out months later, when the customers have already stopped coming.