Skip to content

Pages to Expand

The “Pages to Expand” card on your dashboard surfaces the pages on your site that don’t yet have enough content depth to be quotable by AI engines.

Every page MeetGEO scans gets a depth score from 0–100, calculated from three signals:

  • Word count (0–40 points) — scales up to 1,200+ words
  • Headings (0–30 points) — scales up to 6+ H2 sections
  • Opening paragraph strength (0–30 points) — length of the first real paragraph, up to 80+ characters

This is intentionally simple and explainable. If a customer sees “your About page scored 45/100,” they can reason about what to improve.

ScoreLabelWhat it means
80–100ComprehensiveDeep, well-structured page. Excellent quotation surface.
60–80SubstantialStrong content, well-developed. Good quotation surface.
40–60ModerateAdequate. Could go deeper.
20–40ThinNeeds more depth to be quotable by LLMs.
0–20EmptyAlmost no content. Often a placeholder page.

LLMs prefer to cite sources with depth and structure for two reasons:

  1. Confidence — when an LLM is constructing an answer, it needs sources that contain enough context to support a quotable claim. A 50-word page is harder to cite confidently than a 1,000-word page covering the same topic in depth.
  2. Specificity — H2 sections and a strong lede signal “this page covers a specific topic in a structured way.” That’s exactly the shape AI engines look for when they need to attribute a fact.

A site with 50 thin pages and 5 substantial pages will have its 5 substantial pages cited and the 50 thin ones invisible.

Click the page row in “Pages to Expand” — MeetGEO opens an AI-generated outline grounded in your brand identity. The outline gives you:

  • Section headings (H2s) that match what LLMs expect for your page type
  • Talking points per section, written in your brand’s voice
  • Tone guidance for the lede

Copy the outline into your CMS, expand the page, publish, then click ↻ Re-scan on the row.

Re-scan re-fetches the page right now (instead of waiting for the next weekly update) and re-measures content depth. It updates the score and badge in ~3 seconds.

What it does:

  • Re-fetches the page using a full browser (works on dynamic pages that load content after the initial page load)
  • Recomputes word count, heading count, and opening paragraph strength
  • Updates the depth score in real time
  • If the score crossed into Moderate or higher, the row drops off the list

What it doesn’t do:

  • Doesn’t update schema (that’s part of the weekly process)
  • Doesn’t call any AI engine
  • Doesn’t affect monitoring data

It’s a lightweight content remeasurement only.

A useful baseline:

  • At least 600 words for most page types
  • At least 3 H2 sections organizing the content
  • A first paragraph of 80+ characters that summarizes the page’s core claim

Going past Comprehensive (80+) has diminishing returns. The goal isn’t “longest page possible” — it’s “every important page is at least Substantial.”