AIVO Edge™
Case Study

A global skincare brand was present in every AI response. It was never recommended.

How structured decision testing revealed systematic elimination — and how targeted narrative interventions began reversing it within 8 weeks.
Category: Anti-Ageing Skincare
Brand: Global Mass Premium (Anonymised)
Platforms: ChatGPT · Gemini · Perplexity · Grok
Assessment: 20 Live Runs · 5 Temporal Windows

When consumers ask AI for skincare recommendations, does this brand survive to the final answer?

"
Best anti-ageing skincare brand
"
Dermatologist-recommended wrinkle products
"
Which brand should I choose for visible results?

These are the questions real consumers type into ChatGPT, Gemini, Perplexity, and Grok. The AI system doesn't just answer — it narrows, compares, and resolves. By the third or fourth turn, most brands have been quietly eliminated. The consumer never knows they were an option.

Structured four-turn testing across four AI platforms

4
AI Platforms
ChatGPT · Gemini · Perplexity · Grok
20
Live Conversations
5 per platform · 5 temporal windows
80
Conversational Turns
4 turns per conversation · every state classified
1
Human Analyst Team
Sentiment coding · behavioral classification · pattern identification

Each conversation follows a controlled four-turn decision flow. Our analysts observe every turn in real-time, classifying the brand's state, identifying sentiment shifts, and documenting competitive substitution triggers. Runs are distributed across different temporal windows to capture variation over time.

Present everywhere. Recommended nowhere.

0%
Conversational Survival Rate
Across 20 live conversations on 4 AI platforms, the brand was never the final recommendation. Not once.
P
T0 · Listed
Present in 100% of initial responses
W
T1 · Weakened
Reframed as "accessible" or "starter"
R
T2 · Eliminated
Displaced by "clinically proven" rival
T3 · Not Selected
Never the final recommendation
Conversational Funnel — Brand Survival
T0
Listed
Present among options
100%
T1
Weakened
Reframed as secondary
100%
T2
Eliminated
10%
T3
Not chosen
0%

The brand won the potency argument. It lost the decision.

Our analysts identified a consistent structural pattern across all four platforms. AI systems were resolving purchase decisions using a hierarchy that prioritized safety and clinical authority over efficacy and potency. The decision rule was: Safety > Speed.

How AI systems framed the brand

"Drugstore favorite" · "Budget-friendly" · "Good starter option" · "Accessible anti-ageing"

How AI systems framed the winning competitors

"Dermatologist-recommended" · "Clinically proven" · "Barrier-first formulation" · "Ingredient authority"

The displacement mechanism

The brand was not criticised. It was acknowledged as effective. But when AI systems moved from listing options to making a recommendation, "clinically proven" consistently outweighed "proven effective." The brand was present — but not permitted to anchor the recommendation.

One competitor captured 70% of all displacement decisions

The competitive displacement was not distributed evenly. A single rival brand captured the majority of recommendation decisions, with each AI platform selecting a specific product variant from that competitor's portfolio.

Replacing Brand Displacement Share Platforms
Primary Competitor (various SKUs) 70% All 4 platforms
Derm-positioned brand 10% ChatGPT
Pharmaceutical-heritage brand 10% ChatGPT
Specialist retinoid brands 10% Mixed

The concentration of displacement toward a single competitor is significant. It means remediation can be highly targeted — counter-positioning against one rival, not the entire market.

From 0% to 15% survival in 8 weeks

Working with the client's brand and content teams, we identified two targeted narrative adjustments based on the diagnostic findings. No paid media. No website redesign. No engineering changes.

1
Baseline Assessment0% CSR established
Elimination map delivered
2
Intervention 1Decision-stage positioning
repositioned
3
Intervention 2Attribute reframing in
structured data
4
Re-test (8 weeks)15% CSR measured
Improvement confirmed
5
OngoingMonthly monitoring
continues
Baseline
0%
CSR · Feb 2026
After 2 Cycles
15%
CSR · Apr 2026
+15pp
CSR improvement
-30pp
Replacement rate reduction
-60pp
Best platform improvement
-10pp
Top competitor share diluted

Results from temporal monitoring across two monthly re-test cycles. Individual outcomes vary by category, competitive landscape, and intervention scope. Monitoring is ongoing — target: 25% CSR by month 4.

Being mentioned in AI responses is not enough. If your brand does not survive to the final recommendation, you are present — but not chosen.

— AIVO Edge Assessment Finding

AI recommendation displacement is systematic, measurable, and fixable

This case demonstrates three things that matter for any brand competing in AI-mediated markets:

Displacement is structural, not random

The same brand was eliminated in the same way, at the same conversational turn, across all four AI platforms. This isn't noise — it's a consistent pattern that can be diagnosed and addressed.

Traditional monitoring misses the problem entirely

The brand had 100% presence at Turn 0. Any tool measuring mention frequency would report full visibility. Only multi-turn decision testing reveals that presence without survival is presence without value.

Targeted narrative interventions produce measurable improvement

Two specific adjustments — identified through structured diagnostic analysis — shifted the brand from 0% to 15% survival in 8 weeks. No media spend increase. No website redesign. Structural narrative improvement, validated through controlled re-testing.

See the diagnostic console behind this case study

Explore the live dashboard, temporal monitoring, and evidence vault that produced these findings.