Skip to content
Ranqo
Docs
DashboardGet started
GuidesMethodologyIntegrations
Video tutorials

Metrics

  • How Ranqo measures
  • Visibility
  • Share of Voice
  • Average position
  • Sentiment
  • Citations
  • Source and article types
  • Difficulty and volume

Trends

  • One point per run
  • Carried values
  • Time zones

Measuring impact

  • Measurement windows
  • Noise bands
  • Confirmed wins

Data

  • Competitor identity
  • Why numbers change
  • AI crawlers

6 sections

Metrics

Average position

View as MarkdownOpen this page as .md
ChatGPTOpen in ChatGPTAsk about this pageClaudeOpen in ClaudeAsk about this page

How Ranqo finds your rank inside each AI answer, what the average position covers, and why a lower number is better.

Being named is the first step. Position tells you where you were named: first on the list, or seventh after six competitors.

Average position: Your average rank when AI mentions you. Lower is better.

Definition, in the same words as the dashboard's hover text.

Position in one answer#

When an answer names your brand, a model reads it and records your rank among all the brands and products it names. It applies the first rule that fits the answer:

  1. Numbered list. Your list number is your position: item 3 is position 3.
  2. Bulleted list. The order of the bullets: the second bullet is position 2.
  3. Inline list. The order in a sentence such as "options include A, B and C": C is position 3.
  4. Prose only. The order in which brands first appear in the answer.

If you appear both in an opening summary and in a list, the list position wins. When the model confirms a mention but gives no position, or its reading cannot be used, Ranqo counts the numbered items, bullets, headings, list entries or paragraphs before your first mention instead, and failing those estimates it from how far into the answer the mention comes.

Rule 1 is the one Ranqo sets answers up for. Every engine except Google AI Overviews is asked to use a numbered list when it compares or ranks products, and before an answer is read Ranqo reformats a set of products into a numbered list, with the reformatting told to keep the engine's order. The reformatting sees only the first 8,000 characters of an answer, so a brand named only after that point gets no position and counts as not named. When the reformatted version cannot be used, positions are read from the cleaned answer, in the shape the engine wrote it. See How Ranqo measures and Why numbers change.

What the average covers#

Average position is calculated only over answers that name you with a position. An answer that leaves you out has no position, so it does not count as a bad one: it lowers your visibility instead.

Across engines it is combined the same way as visibility. Ranqo averages your position on each engine, then weights the engines with the same engine weights, rescaled over the engines where you were named with a position. It is shown to one decimal place.

Average position equals sum of your positions divided by answers that name you with a position.

Average position=sum of your positionsanswers that name you with a position
Calculated on each engine separatelyChatGPTPerplexityGeminiGoogle AI OverviewsClaudeGrok

For example, named at positions 1, 3 and 2 on one engine, your average position there is (1 + 3 + 2) / 3 = 2.0.

Lower is better#

Position 1 means you were named first. A falling number is an improvement, so the change arrow on the Position card points up and turns green when your position number goes down.

Read it with visibility#

Because it only covers answers that name you, average position can look strong on very little evidence. A brand named once, at the top of one list, has an average position of 1.0. The Position card on the Visibility page shows how many answers named you out of all the answers in the range, for example "4 of 60 cited". Visibility tells you how often you were named at all.

Position Distribution#

The Position Distribution card on the Visibility page counts every answer in the range, not only the ones that name you:

  • 1st, 2nd and 3rd: answers where you held that position.
  • 4th+: answers where you were named at position 4 or later.
  • Not cited: answers that did not name you.

The Top 3 Rate beneath it is the share of all answers that ranked you in the top three. Unlike average position, it goes down when you are left out, so it cannot look strong on a handful of mentions.

Competitors#

Competitors' positions are read the same way, from the same answers, and averaged with the same engine weights. That makes your position and a competitor's directly comparable on the Competitors page.

Previous page: Share of VoiceNext page: Sentiment
On this page
Ranqo
Docs
Dashboardranqo.aiPrivacyTerms
Be the Source AI Cites.
Ranqo
Docs
GuidesMethodologyIntegrations

Metrics

  • How Ranqo measures
  • Visibility
  • Share of Voice
  • Average position
  • Sentiment
  • Citations
  • Source and article types
  • Difficulty and volume

Trends

  • One point per run
  • Carried values
  • Time zones

Measuring impact

  • Measurement windows
  • Noise bands
  • Confirmed wins

Data

  • Competitor identity
  • Why numbers change
  • AI crawlers
Get started