Definition
A GEO score is a composite number a vendor calculates to summarize how well a site is set up to be cited by AI assistants. Unlike a citation rate, which counts an observed outcome, a GEO score blends observed appearances with page-level checks the vendor chose to weight. No two vendors compute it the same way.
At a glance
- CategoriesMeasurement
- Related fieldsAnalytics
- DifficultyBeginner
At a glance
- A GEO score is a vendor construct, not an industry standard, and there is no specification to check it against.
- A score mixes two unlike things: what assistants actually did, and what a crawler found on your pages.
- The weights are the product, and they are the part a vendor tends not to publish.
- Two vendors scoring the same site can move in opposite directions in the same month.
What goes into a composite, and who decided
Open any GEO score and the inputs fall into two groups.
The observed group comes from running prompts: how often you appeared, how often you were cited, how you sat against competitors. Anyone with the same prompt list can reproduce these.
The inferred group comes from crawling your site: whether pages answer directly, whether structured data is present, whether content renders without JavaScript, whether an author is named. These are proxies, describing conditions associated with being cited rather than citations.
The score multiplies both groups by weights and rolls them into one figure. Nothing in that step is visible from outside. A vendor weighting crawlability heavily will score a clean, quiet site above a messy, frequently cited one, and both numbers look equally authoritative on a slide.
One number is a conversation starter, not a target
A composite is good at one job: getting a room to agree there is a problem. It is poor at the rest.
Optimizing for the score is the failure mode. Once a team knows which checks feed it, the cheap wins come from satisfying checks rather than from being useful to a buyer, and the number rises while the answers stay the same.
Comparison is the second trap. Your score and a rival’s score from a different vendor share a name and nothing else.
Keep the composite for the headline and manage the work on the parts. Share of voice tells you whether you are gaining on rivals. Source credibility signals tell you what to fix on a page.
Interrogate a score before you report it upward
- Ask for the input list. Not the marketing description. The actual metrics feeding the number.
- Ask which inputs are observed and which are inferred. If the split is not clear, the score is mostly a site audit wearing a visibility badge.
- Ask what the prompt set is and who chose it. A score built on prompts you already win is a compliment, not a measurement.
- Ask when the weights last changed. Then check your history for jumps that line up with product releases rather than your work.
- Track two components beside it. One observed metric, one you can check by hand. When they diverge from the composite, trust them.
What sits under a typical GEO score
| Input | Observed or inferred | Can you verify it yourself |
|---|---|---|
| Appearance across a prompt set | Observed | Yes, by rerunning the prompts |
| Citation of your URLs | Observed | Yes, by reading the source lists |
| Answer-first page structure | Inferred | Yes, by reading your own pages |
| Structured data coverage | Inferred | Yes, with any validator |
| Overall weighting | Vendor decision | No |
Frequently asked questions
-
Is a GEO score the same as a visibility score?
They overlap heavily and the names get used loosely. A visibility score leans on observed appearances in answers. A GEO score usually folds in a site readiness audit too. Neither is standardized, so the label tells you less than the input list does.
-
Can I compare my GEO score to a competitor's?
Only if both came from the same vendor, the same month, the same prompt set. Otherwise you are comparing two formulas that share a word. Competitive comparison is useful, but it has to run inside one measurement system rather than being assembled from two marketing pages.
-
Is a GEO score worth paying for?
For a small site, not yet. You can reproduce the observed half with a prompt list and a spreadsheet, and the inferred half by reading your own templates for an afternoon. A score earns its cost once you have enough pages and prompts that doing it by hand stops working.
-
Why did my score drop when I did nothing wrong?
Three usual causes: the vendor changed its weights, an assistant changed how it cites, or a competitor displaced you on shared prompts. Only the third is about you. Check whether your observed metrics moved first, because a composite can fall while every measurement under it holds steady.