Every AI visibility program eventually has to answer one question honestly: are we chasing mention volume, or are we chasing recommendation quality? They correlate, but not tightly enough to treat as one metric — and our 137-brand insurance benchmark found the gap between them is where most of the strategic insight actually lives.
Being Named a Lot Is Not the Same as Being Recommended
State Farm leads the Q2 2026 leaderboard by a wide margin: 457 organic mentions, the most raw citations (129), and the highest Top-3 rate among the top 10 (79.9%). But mention volume converted to active recommendation language only 40.9% of the time. USAA, with 187 fewer mentions (270 total), converted 46.3% of its mentions into active recommendations — the highest recommendation rate in the top 10.
The pattern holds across the leaderboard: recommendation rate does not track cleanly with mention count. Nationwide (5th by mentions, 252 total) converts 41.3% — almost identical to State Farm's 40.9%, despite 205 fewer mentions. Allstate (6th, 276 mentions) converts 37.7%. Being named constantly builds awareness; it does not, by itself, buy endorsement.
The Prominence Outlier: When Rank Doesn't Predict Placement
The clearest volume/quality split in the study isn't about recommendation rate at all — it's about where a brand lands when it appears. The Hartford ranks 8th overall by mention volume (134 mentions) but posts the single highest Top-3 placement rate of any top-10 brand: 88.1%, ahead of even State Farm's 79.9%. When The Hartford shows up in an AI answer, it is almost always surfaced near the top — despite ranking eighth on raw mention volume.
Why this happens: mention volume is often inherited from category-wide presence and long-standing scale. Prominence and recommendation language require the engine to have a specific, citable reason to feature or endorse a brand — a strong Top-3 answer position, a sourced comparison, transparent claims. Brands with volume but no fresh citable proof get named without being championed or prominently placed.
The Real Range: 27.6% to 46.3%
Across the top 10 brands in the Q2 2026 benchmark, recommendation rate ranges from The Hartford's 27.6% (the lowest, despite its Top-3-rate lead) to USAA's 46.3% (the highest). That's the real spread — and it shows recommendation rate and prominence can move in opposite directions for the same brand.
| Brand | AIVI™ score | Mentions | Top-3 rate | Recommendation rate |
|---|---|---|---|---|
| State Farm | 100.0 | 457 | 79.9% | 40.9% |
| Progressive | 71.2 | 308 | 58.4% | 41.2% |
| USAA | 69.1 | 270 | 56.3% | 46.3% |
| GEICO | 65.6 | 265 | 75.1% | 42.6% |
| Nationwide | 63.4 | 252 | 46.4% | 41.3% |
| Allstate | 61.6 | 276 | 55.8% | 37.7% |
| Travelers | 46.0 | 170 | 59.4% | 34.1% |
| The Hartford | 42.0 | 134 | 88.1% | 27.6% |
| Chubb | 41.2 | 140 | 60.7% | 32.1% |
| Amica | 39.9 | 124 | 66.9% | 39.5% |
Source: Brainpan.AI Insurance AI Visibility Index™ — Q2 2026 benchmark, 137 brands, Finding 01 (Overall Leaderboard).
Which One Should You Optimize For?
It depends on where the brand already sits. A brand with low mention volume has a discovery problem — it needs broader category presence before recommendation quality even has a chance to matter, because the engine has to name it before it can endorse it. A brand with solid volume but a below-average recommendation rate has a trust problem: more mentions alone won't fix it, but supplying the citable proof engines reward (ratings, transparent pricing, satisfaction data) can shift existing mentions from passive to active recommendations without any additional media spend.
Frequently Asked Questions
What's the difference between mention volume and recommendation quality in AI visibility?
Mention volume is how often an AI engine names a brand across responses. Recommendation quality is how often, out of those mentions, the engine actually endorses the brand with active recommendation language. A brand can have high volume and comparatively lower conversion, or the reverse.
Can a brand have high mention volume but a lower recommendation rate than a smaller competitor?
Yes. In our Q2 2026 benchmark, the volume leader (457 mentions) converted 40.9% of mentions into active recommendations, while a competitor with 187 fewer mentions converted 46.3% — the highest recommendation rate in the top 10. Being named most often doesn't guarantee the highest endorsement rate.
Is it better to chase more mentions or a higher recommendation rate?
It depends on your starting point. If a brand rarely gets named, it needs broader category presence before conversion optimization matters. If a brand is already named often but converts a lower share of those mentions, supplying citable proof points to lift recommendation language is the faster lever.
What is a "prominence outlier" in AI visibility?
A brand that ranks lower on overall mention volume but appears disproportionately near the top of the answer when it does show up. In our benchmark, one carrier ranked 8th by mention volume but posted the single highest Top-3 placement rate of any top-10 brand — 88.1%, well above the leader's 79.9%.
How is recommendation rate measured?
In our methodology, every response naming a brand was scored for whether the surrounding language constituted an active recommendation versus a neutral mention with no endorsement. Recommendation rate is the share of a brand's total mentions that included active recommendation language.
Find out if you're named or recommended
Get a benchmark of your brand's mentions, citations, and recommendation rate against the competitors that matter, across all five major AI engines.
Request AI Visibility AuditPrefer to browse first? Download a sample audit (PDF) →
Kevin Walsh
Kevin Walsh is the founder of Brainpan.AI, where he builds AI visibility infrastructure, GEO/AEO strategy, schema systems, and citation optimization programs for brands that need to be retrieved, cited, and trusted by AI answer engines.