The GEO Data Trust Problem: SEOs Want the Data, Not the Platforms

GEO (Generative Engine Optimization) has become SEO's hottest topic lately, but trust in the tools built to measure it remains an open question. Duane Forrester, founder and CEO of UnboundAnswers.com, ran a three-week survey in July among practitioners working on visibility in AI search. It drew 163 responses, sourced through his own network, shares, and paid promotion on LinkedIn and X. Because the sample is self-selected, it reflects deeply engaged practitioners rather than the industry at large; the margin of error is roughly 7%.

The survey hinges on two numbers. Asked to rate the value of five types of AI visibility data — query adaptation beyond keywords, competitive comparison, determining whether a brand mention came from model training or real-time search, snippet-level attribution, and citation status — respondents gave an average score of 4.20 out of 5. But asked how sensible it is to invest in a standalone platform delivering that data, the score dropped to 3.19. Only 44% of respondents considered buying such a tool worthwhile, while nearly a third rated it a 1 or 2 out of 5.

123 people (75%) wrote detailed answers to an open-ended question, revealing deep-seated distrust. The most common themes: opaque, untrustworthy methodology (24%), inability to tie results to business value / ROI (20%), model instability and personalized responses (15%), artificial query lists that don't reflect real user demand (11%), and data that doesn't translate into action (9%). Several respondents described the same problem: you build the list of queries you track visibility against yourself — meaning you decide in advance what you should be visible for, then grade yourself against your own list; one respondent called it a "self-fulfilling prophecy."

The key finding is that the problem isn't price. Only 7% of respondents cited cost as an issue, while 57% said either "I don't trust the number" or "I can't tie it to money." Forrester calls this the survey's most useful takeaway: what's holding back this category of tools isn't their cost — it's distrust of them. Deeper analysis shows that respondents who raised trust concerns value the data just as highly as everyone else (4.20 vs. 4.19) and have comparable budgets, but their willingness to invest in a platform drops from 3.36 to 2.76 — meaning the issue isn't money, it's trust.

The survey also revealed exactly what data respondents want: query adaptation beyond keywords (90%), competitive comparison (83%), knowing whether data came from model training or real-time search (83%), snippet-level attribution (75%), and citation status (71%). As for which systems matter most, Google's AI Overviews and AI Mode led at 95%, followed by ChatGPT at 94%, Gemini at 75%, Claude at 64%, Perplexity at 34%, and Copilot at 25%.

Forrester puts the results in broader context: according to IBISWorld, roughly 715,000 people work in SEO and internet marketing consulting in the U.S. alone, yet only 163 of them — one in every 4,400 — answered the survey. He suggests that gap between how often the industry discusses AI search as an existential threat and the actual level of hands-on engagement may be the survey's most honest conclusion: plenty of talk, not much practical action.