THE AIVO METHOD › Which Metrics Hold Up
Not Every AI Visibility Metric Deserves a Decision
Some AI visibility metrics carry a confidence interval. Others do not. Knowing the difference determines what you can act on.
Rates vs facts
Some AI visibility metrics are rates that carry a 95% confidence range: presence, citation rate, and sentiment as a percent are proportions and get an interval. Rank position is the noisiest metric in AI visibility; AIVO converts it into top-1 rate and top-3 rate, which are clean proportions.
Carries a 95% confidence range
Reported differently
Carries a 95% confidence range
Presence / visibility rate
Appeared in k of n runs
Sentiment as a continuous score
Mean-based interval, not a proportion
Recommendation / inclusion rate
Proportion
Accuracy as a single confirmed error
A fact, not an estimate
Citation rate
Proportion
Rank or position
Ordinal, highly unstable
Share of answer / share of voice
Ratio of counts, with care
Severity of a finding
Human judgment, scored not estimated
Sentiment as percent positive
Proportion
Raw catalog facts
A record, not an estimate
Accuracy as recurrence
Proportion
Reported differently
Sentiment as a continuous score
Mean-based interval, not a proportion
Accuracy as a single confirmed error
A fact, not an estimate
Rank or position
Ordinal, highly unstable
Severity of a finding
Human judgment, scored not estimated
Raw catalog facts
A record, not an estimate
Each proportion is a rate drawn from a sample, so the true rate could sit anywhere across a known range given that sample size. AIVO reports that range as a 95% confidence interval rather than a single percentage.
Why rank is the trap
Rank position is the most frequently cited AI visibility metric and the least stable one. A tool that reports a single fixed AI rank for a brand is reporting a number that shifts on its own, run to run, independent of anything the brand does. AIVO decomposes rank into threshold proportions instead: top-1 rate (the share of runs where a brand appears first) and top-3 rate (the share where it appears in the top three). Each is a clean binomial proportion with a 95% confidence range.
WHAT MOST TOOLS REPORT
#3
A single rank that shifts every run.
WHAT AIVO REPORTS
Top-1 rate: 12% (6% to 22%)
Top-3 rate: 48% (38% to 58%)
Each is a proportion with a 95% confidence range.
Illustrative example.
STABILITY
The stability heuristic
Presence
comparatively stable. Trust it.
Rank position
close to noise. Treat any single reading as uninformative.
Sentiment and comparison
directionally useful, volatile on any one reading. Trend them.
Never pool across platforms
ChatGPT, Claude, Google AI, and Perplexity are not exchangeable. Each has a different retrieval process, training data, and update cadence. AIVO estimates every metric within a single platform, compares those estimates across platforms side by side, and never blends them into one headline number.
ChatGPT
62%
48–74%
Presence (95% range)
Claude
41%
29–54%
Presence (95% range)
Perplexity
55%
42–67%
Presence (95% range)
Illustrative. Same brand, same prompts — different platforms.
Know what AI says about you.
And how sure you can be.
Independent measurement of how AI represents your brand. Decision-grade intelligence, one market, under 30 days.
30-minute scoping call · no commitment