How to audit your own website in an hour

Titles and descriptions, ten minutes

Open each page and look at the browser tab. That is the title. It should name the page's job, not just your company, and it should be short enough to show whole in a search result. Then view the page source and find the meta description. It should exist and it should be written for that page.

The usual finds: the same title on every page, a description missing, or a title so long it gets cut off. Write down each page that fails. You will fix them after the audit, not during, or the hour becomes a day.

One main heading and honest subheadings, five minutes

Each page needs exactly one main heading, and it should match what the title promised. Look for pages with none, and for pages with two or three competing for the job. Then read the subheadings alone, top to bottom. They should tell the story of the page without the paragraphs.

Links, sitemap and canonical, fifteen minutes

Click every link in your navigation and footer, then the links inside your main pages. A broken link is a dead end for a reader and a sign of neglect to a crawler. Write each one down.

Then load your sitemap by adding sitemap.xml after your domain. Every address in it should open a real page, and every real page should be in it. Finally, view the source of a few pages and find the canonical link. It should point to the page's own clean address, not to a version with tracking codes on the end.

• Navigation and footer links all open.
• The sitemap lists real pages and nothing else.
• The canonical link points to the clean address of the same page.

Mobile, the not-found page and contrast, ten minutes

Open the site on your own phone. Read a whole page. Tap every button with a thumb. Text you have to pinch, buttons that sit too close together, and a menu that will not open are all fails.

Then type a made-up address on your domain. You should see a branded not-found page, and it should return a real 404 status, not a 200 that pretends the page exists. Your browser's developer tools show the status in the Network tab.

Last, contrast. Grey text on a light background is the common failure. Your browser's accessibility panel flags text that does not pass WCAG 2.2, or paste the two colors into any contrast checker.

The form and the structured data, ten minutes

Fill in your own contact form and send it. Then check where it should have landed: your inbox, your CRM, your phone. A form that looks fine and delivers nowhere is the costliest fail on this list, and it is common.

Then paste a page address into Google's Rich Results Test. It reads the structured data on the page and reports what parses and what does not. An error there means the search engine is guessing about who you are.

What to do with the list, the last ten minutes

Sort the fails by the reader. Anything that stops a person from contacting you goes first: the form, the phone view, broken links. Titles and descriptions go second. Structured data and canonical tags go last, because a reader never sees them.

Set a reminder to run the audit again in a few months. Sites drift. A plugin update, a new page, a changed phone number, and something on this list breaks quietly.

How the builder's check relates to this list

Website Builder Studio has a check that runs before anything publishes, and it covers most of this list on every page. Each page gets a title and description, one main heading, a clean address, a sitemap entry, breadcrumbs, structured data, and a branded not-found page with a real 404 status. The check is validated against Google Search Essentials and modern web standards. A page that fails does not publish, and the in-app assistant says why.

The form is covered too. Connecting a CRM ends with a Send test lead button that proves the delivery before a real lead flows. What the check cannot judge is whether the words are genuinely useful. That part of the audit is still a person reading a page and asking whether it answers a real question.