How do I measure GEO visibility and ROI?

Measuring GEO visibility is simple in concept: run the same questions before and after, score the answers the same way, and see what moved. ROI is harder because a mention does not always come with a click. The practical approach is to value a mention based on what one new customer is worth, then compare that to the cost of the build.

The visibility score

Pick 20–30 questions. Run each through ChatGPT, Claude, Perplexity, and Google AI Overviews. Score each result:

• — Whether the business is mentioned and whether it is clearly recommended.
• — Which competitors are recommended instead for each question.
• — Whether the business facts shown are accurate and supported by strong sources.

The score summarizes the same observed signals across the full question set. Keep the questions and evaluation rules consistent between checks; the evidence and direction of change matter more than the absolute number.

What to track alongside the score

• — Which engine moved and which did not.
• — Which questions moved and which stayed flat.
• — Whether the answer is correct, wrong, or outdated.
• — Whether an AI Overview appeared for each query.

How to think about ROI

Work out the contribution from one customer, not just the sales amount. Compare that with Found + Follow-up at $950 to start, then $299 a month. Everything included. Getting found and follow-up are automated together, but no number of jobs or return is guaranteed.

What not to measure

• — Keyword rank alone. It says nothing about ChatGPT or Perplexity.
• — Total impressions from Search Console. They include surfaces GEO does not touch.
• — Clicks from AI citations. Most engines do not pass click data.

Sources

Search Console documents what Google reports about AI feature clicks. Engine-specific behavior must be observed manually; vendors do not publish full citation logs.