I checked ten pricing pages. Stripe has 133 h1 tags.

I ran our page checker over ten pricing pages from companies whose engineering nobody questions, expecting a boring afternoon and a post about how the professionals do it properly.

Stripe’s pricing page has 133 h1 tags.

Not a typo. One hundred and thirty-three. The specification allows it and browsers render it fine, so nothing is broken in any user-facing sense, but the h1 is meant to say what the page is about, and a page making that claim a hundred and thirty-three times is making no claim at all. Zapier has two, which is the ordinary version of the same mistake. Everyone else has one.

That was the first surprise. The second was structured data.

Five of the ten pages publish none whatsoever. Notion, Vercel, Slack, Asana and Airtable all serve a pricing page with no schema markup on it, which means no Product or Offer information, and no possibility of a rich result. Stripe, Linear, Figma, Zapier and monday.com each have some.

I had assumed this was solved. It is not solved.

Then I looked at length and stopped assuming anything. Linear’s pricing page runs 433 words. monday.com runs 8,081. Same page type, same job, nineteen times the text. Both scored well on the check and both are perfectly reasonable pages, which tells you something useful: there is no correct length for a pricing page, and anybody quoting you one is repeating a number they read somewhere.

Internal links spread almost as wide. Airtable points at 18 pages from its pricing page. Stripe points at 300.

Three hundred internal links from one page is not an error either. It is a decision about what a pricing page is for, and Stripe has decided theirs is a hub. Airtable has decided theirs is a destination. You can defend both, and you cannot defend having never thought about it, which is the state most pricing pages are actually in.

Here is what the exercise taught me about audits, which is really what this is about.

A checker gives you facts, not verdicts. It can tell you there are 133 h1 tags. It cannot tell you whether that matters for Stripe, whose brand search volume means they will rank for their own pricing page regardless of how the headings are marked up. The finding is real, the consequence depends entirely on who owns the page, and every audit that reports issues without that second half is producing a list nobody will action.

The thing worth stealing is the shape of the check. Status, indexability, canonical, title length, one h1, heading order, internal links, structured data, word count. Nine things. They cover most of what actually goes wrong, and you can look at all nine on a page in about two minutes without any tool at all.

What I keep turning over is Asana. Their page scored lowest of the ten, at 84, on a title of 75 characters and missing structured data. Seven thousand words of carefully written pricing content, and the two things holding the score down would take somebody twenty minutes on a Tuesday. Nobody has done it, and I suspect nobody has done it because nobody owns the page in a way that includes the head of the document.

Questions people ask about this