Nike Scores 100 on AI Mentions. It's Losing on Pricing Narrative. Here's What That Means for Your Brand.

Nike scores a perfect 100 on AI brand mention frequency. Ask any AI assistant about running shoes, training gear, or athletic footwear and Nike appears — not sometimes, virtually always. By the measure most marketing teams are using to track AI visibility, Nike is the dominant brand in its category. And yet, over six consecutive quarters through 2025 and into 2026, Nike's revenue fell. In fiscal 2025 alone, sales dropped 10% — the largest annual decline since 1987. The mention count was perfect. The business was contracting. Something in between those two facts is doing damage that standard AI monitoring can't see.

That something is the pricing narrative. Layer underneath the mention frequency and AI consistently surfaces the same pattern: Nike is described as "expensive," "overpriced relative to newer competitors," and "the mainstream choice, but not the best value." Competitors On, Hoka, and Salomon are routinely positioned by AI as fresher, more innovative, and more worth the money. Adidas gets described as offering comparable quality at lower price points. Nike appears first, and then AI immediately qualifies that recommendation with a pricing caveat that lands on the buyer as a red flag.

The Gap Between Mention Count and Narrative

The instinct in most marketing orgs, when told their brand scores perfectly on AI mentions, is to move on. The problem is framed as solved. But mention count and narrative are two distinct layers of AI brand presence, and a brand can dominate one while losing on the other.

Mention count tells you: does AI know your brand exists? Narrative tells you: what does AI say about your brand when it brings it up? For Nike, the narrative layer carries consistent pricing damage. The brand appears — and then AI hedges. It notes that newer entrants offer comparable performance. It surfaces the China pricing concession (Nike was forced to compete on price in China to defend market share, and that signal was absorbed into training data and now appears in global AI responses as evidence of pricing pressure). It positions Nike's mainstream status as a liability rather than a strength when budget-conscious buyers are asking for recommendations.

The result: AI is recommending Nike with a caveat the buyer hears as a reason not to buy. The mention gets logged as a win. The lost sale never gets attributed to anything.

"AI narrative bleed is when a brand is leaking sales it can't see or attribute — buyers are being talked out of a purchase by AI, in a channel the brand has no visibility into."

Coining the Term: AI Narrative Bleed

Nike is the cleanest live example of AI narrative bleed at scale. Its NPS remained at +43 as recently as February 2025 — measured by Citi — meaning brand love among existing customers is intact. RepTrak recorded a 7-point drop in Nike's corporate reputation score in late 2024, which suggests the reputational signal was moving. But NPS and reputation scores don't tell you what AI is saying about your brand when a new buyer, with no existing relationship to Nike, asks a question in ChatGPT or Perplexity about what running shoe to buy.

That buyer gets: Nike mentioned first (high mention score), followed by language about price sensitivity, followed by a comparison to On or Hoka that positions those brands as the smarter choice for someone who cares about value. The AI didn't recommend against Nike. It recommended Nike and then made the competitor recommendation feel safer. That's the bleed.

Why Standard Monitoring Misses This

A April 2026 analysis by DerivateX identified an inverse version of this problem: companies with perfect sentiment scores can still have mention frequency problems that suppress brand visibility. Nike's situation is the reverse — maximum frequency with narrative damage running underneath it. Both failure modes are invisible to tools that track only one dimension.

The monitoring gap has a structural cause. Most AI brand tracking approaches were adapted from social listening and SEO monitoring — tools designed to count appearances and flag sentiment in isolation. They were not built to track the relationship between those two signals, or to surface the specific narrative context in which a brand appears. A brand that appears in 100% of responses about its category but is consistently described as "the expensive option" does not look broken in a mention-count dashboard. It looks like the category leader. The damage only becomes visible when you read what the AI actually says.

• Nike appears in virtually every AI response about athletic footwear — mention frequency is statistically dominant.
• AI responses consistently attach pricing qualifiers: "expensive," "premium-priced," "better value alternatives exist."
• On, Hoka, and Salomon are positioned by AI as "more innovative" and "better value" — AI is doing competitor positioning against Nike at the point of buyer research.
• The China price-competition signal — absorbed into AI training data — now surfaces in global narrative as evidence Nike is under pricing pressure.
• No standard marketing report captures any of this. The lost sales register as unexplained churn or category softness.

The Allbirds Warning

Nike's situation is AI narrative bleed at the scale of a $30B brand. The terminal version of the same dynamic played out with Allbirds. At its 2021 IPO, Allbirds was valued at $4B and AI narratives were uniformly aligned with "sustainable innovation" and "the next Nike." By 2026, the company sold for $39 million. The AI narrative had quietly shifted — from "sustainable innovator" to "overhyped and overpriced" — and Allbirds had no mechanism to detect the shift happening, or to understand how much of their conversion problem was being manufactured upstream of any channel they could measure.

The scale is different. The mechanism is identical. A brand's AI narrative shifts. The brand can't see it. Sales fall. The attribution problem means no one connects the cause to the effect until the damage is done.

What AI Brand Narrative Tracking Actually Requires

Catching AI narrative bleed before it compounds requires monitoring that operates at the narrative layer, not just the mention layer. Mention count is a valid and useful signal — it tells you whether AI knows your brand. The question narrative tracking answers is different: when AI brings up your brand, what does it say, and how does that compare to what it says about your competitors?

For Nike, a proper AI brand narrative tracker would have surfaced the pricing complaint pattern well before six quarters of declining revenue confirmed the problem. The signal was in the AI responses. The monitoring infrastructure to read it wasn't there.

• Track the narrative layer separately from mention frequency — they move independently and mean different things.
• Monitor pricing language specifically — AI pricing framing is the highest-conversion-impact narrative element for most consumer brands.
• Run competitor narrative comparisons — if AI is positioning On or Hoka as better value in the same response that mentions your brand, that's an active suppression event, not background noise.
• Watch for training data lag artifacts — the China pricing signal is in Nike's AI narrative because it was absorbed from news coverage. Identifying those artifacts lets you decide whether to address them with new content or ignore them.
• Treat 73% buyer pre-research as the baseline — if the majority of buyers are using AI before they decide, the AI narrative is the first sales conversation your brand is in. It should be monitored with the same rigor as any other sales channel.

Nike's mention score is 100. Its narrative score is doing damage that took six quarters and a 10% revenue decline to show up in financial results. By the time it's visible in earnings, the AI narrative bleed has already done its work. The gap between those two numbers — maximum visibility, damaged narrative — is exactly what AI brand narrative tracking is built to close.