Recommendation Intelligence Research™ · Study #25

Give the model the one fact it's missing. The brand goes from invisible to everywhere.

The last study in this series showed that a model can know six facts about a brand and use, on average, just one, most of what's stored never gets said. This is the causal follow-up: what happens if you stop waiting for the model to remember, and instead put one real, verified fact directly in front of it, as if a retrieval step had just succeeded. Across 4 brands that barely get recommended today, doing that raised the mention rate by an average of 77.9 percentage points. But the size of that lift, and how much of it comes from the model actually citing the fact versus just being nudged to say the name, differs brand by brand in a way that matters.

77.9ppAverage mention-rate lift
4Underperforming brands tested
50.4%Avg. fact actually cited, when mentioned
1,050New calls, incl. judge scoring
Where this comes from

Stop waiting for memory. Just hand it the fact

Possession vs Deployment measured a passive gap: facts a model already has in memory, and how often those facts show up unprompted. This study manipulates the same mechanism directly. Four brands were chosen specifically because they barely get recommended today, real baseline recommend rates from 0% to 16.5% in the published Recommendation Reports. For each, one real, verified, distinctive fact was identified and injected as a system message reading "Additional context retrieved for this query: {brand}: {fact}", simulating the moment a retrieval step has just succeeded, then the model was asked the brand's own already-published real buyer questions, unmodified, and the response was checked for whether the brand appears at all, and, on a capped sample of the cells where it does, whether the specific injected fact is actually cited or paraphrased, not just the brand name dropped in.

The baseline side of the comparison is entirely reused, the same already-published 20-prompt buyer-question data behind each brand's Recommendation Report. Only the injected condition (10 repeats per prompt) and the fact-usage scoring are new.

Model: gpt-4o throughout
Brands: 4, deliberately chosen underperformers
Injected calls: 800 (4 brands × 20 prompts × 10 repeats)
Judge calls: 240 (capped 60/brand, mentioned cells only)
Baseline data: reused, published Recommendation Reports
New calls total: ~1,050
Why "additional context retrieved." This phrasing simulates a retrieval step that has already succeeded, it is not a claim about how often real retrieval actually surfaces this fact in practice. The question this study answers is narrower and more useful: once the fact is in front of the model, does that fix the deployment side of the gap the last study measured? Brand identities are anonymized throughout this page (Brand A-D) to protect the individual businesses studied.
The finding

Every brand jumps. Not by the same amount

Three of the four brands went from barely-mentioned to near-universal: 1.25% to 97.5%, 5.75% to 97%, and 16.5% to 96%. The fourth, Brand D, the one with a true 0% baseline, moved much less, to 44.5%. That's still a real lift, but it's the smallest in the cohort by a wide margin, and the likely reason is visible in the response text itself: Brand D's injected fact is less directly relevant to the specific buyer questions it was tested against than the other three brands' facts, so the model is less willing to work it into an on-topic answer. A retrieved fact only helps as much as it actually fits the question being asked.

Mention rate, baseline vs. with fact injected
4 brands · gray dot is the published baseline, orange dot is with the fact injected · brands anonymized
Baseline, published Recommendation Reports
With fact injected, this study
The diagnostic

Mentions rise faster than fact citations

A bigger question than the lift itself: when the brand gets mentioned, is the model actually using the fact it was handed, or just nudged to say the name? On a capped sample of mentioned cells, a judge call checked whether the specific injected fact (or a clear paraphrase) shows up in the response. The lift-to-fact-usage ratio, mention-rate lift divided by fact-usage rate, is the diagnostic: a ratio near 1 means the brand only gets mentioned about as often as the fact gets cited, a tight coupling. A ratio well above 1 means mentions are outrunning fact citations, the brand is getting named more from the framing alone than from anything specific being said about it.

Fact-usage rate by brand, when mentioned
Share of judged mentioned cells where the injected fact was actually cited or paraphrased · tag shows lift-to-fact-usage ratio
Brand A · n=60 judged
88.3%
Ratio 1.09, tightest coupling
Brand B · n=60 judged
46.7%
Ratio 1.96
Brand C · n=60 judged
33.3%
Ratio 2.39, widest gap
Brand D · n=60 judged
33.3%
Ratio 1.34
Every brand in this cohort has a ratio above 1. None showed fact-usage keeping pace with the mention-rate lift, which means the effect measured here is, across the board, more about getting the brand into the context window at all than about the model reciting the specific fact back. Brand A comes closest to a 1:1 coupling, when it's mentioned, it cites the fact 88.3% of the time. Brand C shows the widest gap, mentioned readily but the specific fact shows up in only a third of those mentions, the framing itself is doing most of the work.
Why it matters

This is the fix for the gap the last study found

Possession vs Deployment showed the bottleneck isn't whether a model knows something about a brand, it's whether that knowledge reaches the specific sentence being generated. This study shows that bottleneck is fixable, at least in the narrow sense tested here: put the fact where the model can see it at the moment of the answer, and mention rate moves dramatically, even for brands that are functionally invisible today. That's the practical version of AIVO's Linkage Gap versus Reasoning Gap distinction, a Linkage Gap is fixable by getting information into context; a Reasoning Gap, where the model has the information and still doesn't act on it, is not. Every brand here showed lift outpacing fact-usage, the signature of a Linkage Gap being closed by simple presence in context, not a structural reasoning failure being overcome.

What this doesn't prove

"Retrieved" is simulated, not measured

Stating the limits up front. This study injects the fact directly, it does not measure how often a real retrieval system would actually surface this exact fact for this exact query, that's a separate, harder question this doesn't answer. The judge that scores fact-usage runs on the same model that generated the responses, a self-grading risk. The mitigation is a manual spot-check: 15 judge calls checked by hand against the raw response text. 14 of 15 matched. The one disagreement was a likely false negative, a response that closely paraphrased the injected fact but wasn't marked as using it, which if anything means the fact-usage rates reported here are a slight underestimate, not an inflated one.

Four brands, deliberately selected as underperformers, not a representative cross-brand sample, this is a floor-case study by design. 10 repeats per prompt in the injected condition versus 20 in the reused baseline, the baseline's CI is already tight from the larger published sample. One category mix, consumer ecommerce, one model. Cross-platform and cross-model versions of this question are a separate, larger, and currently API-key-blocked follow-up.

Supporting evidence

Two checks that the setup was clean

14 of 15 Manual spot-check agreement

15 judge calls checked by hand against the raw response text, spanning all 4 brands. 14 matched exactly. The one disagreement was a conservative false negative, a close paraphrase marked as not using the fact, biasing the reported fact-usage rates slightly downward, not upward.

4 of 4 Brands show ratio > 1

Every brand tested showed mention-rate lift outpacing fact-usage rate, a consistent Linkage Gap signature across the whole cohort, not a result driven by one outlier brand.

Find out what facts your brand isn't getting credit for

Free AI Commerce Score™ in 10 seconds.

If putting the right fact in front of a model can move mention rate this much, the fastest way to find out which facts about your brand are missing at the moment it matters is a free scan.

Free · No signup · Results in 10 seconds
Keep reading

The rest of the research series

This is the causal half of the possession-deployment question, and sits next to the within-conversation displacement result measured one study earlier.