# Answer-ready sections

> Part of [Lensy by Perseverance AI](https://perseveranceai.com/). The index of every page for AI agents is at [https://perseveranceai.com/llms.txt](https://perseveranceai.com/llms.txt).

Last updated: 2026-11-12

AI search check S5, worth 15 points.

## What it measures and why

On the HTML: one H1 4 (more than one 2). No heading skips 4 (one skip 2), counted only with at least 2 headings. Median words per H2/H3 section: 40 to 500 7; under 40 or 501 to 900 4; over 900 1; no sections 0.

AI search retrieves passages of about a heading plus its paragraphs, so each section should answer one question on its own.

Evidence:

- [Passage-level retrieval (Paz.ai)](https://www.paz.ai/glossary/passage-level-retrieval)

## How it counts

S5 is one of six AI search checks, worth 15 of the score's 100 points. It gives full, partial or no points, and the AI search score is the sum of the six. A page that can't be reached at all gets no score; the reachability line above both scores says why.

## Check it yourself

Scan the page on the [home page](https://perseveranceai.com/#scan), or run `lensy check <url>` from the [command line tool](https://perseveranceai.com/docs/cli) and look for `S5` in the report. A failed check comes with the fix worth the most points first. `lensy explain S5` prints what it measures, why it matters and the evidence in your terminal, and [every check](https://perseveranceai.com/docs/checks) lists the rest.

[All checks](https://perseveranceai.com/docs/checks)
