As generative AI quietly became a front-door shopping assistant in 2025, one thing became clear: throwing money at ads didn’t move the needle. Content did.

That’s the central takeaway from Bluefish’s 2025 Holiday AI Commerce Report, one of the first comprehensive analyses of how AI-driven shopping actually played out during the December holiday rush. By examining millions of AI-generated answers across major platforms, the AI marketing firm mapped which brands, publishers, and narratives shaped what consumers saw when they asked AI systems what to buy—and where to buy it.

The results should unsettle anyone still relying on paid media as a holiday growth lever.

AI assistants largely ignored ads

According to Bluefish, paid media had little direct influence on AI-generated shopping recommendations throughout the holiday season. In fact, the findings echo earlier industry research showing that up to 95% of AI citations came from non-paid sources.

Instead of rewarding ad spend, AI assistants consistently favored brands with:

• High-quality, clearly structured content

• Strong organic visibility

• Consistent messaging across owned and earned media

In practical terms, AI visibility emerged as a leading indicator of demand capture during the most competitive sales window of the year. Brands that weren’t visible to AI simply weren’t part of the conversation—no matter how much they spent elsewhere.

“Holiday 2025 proved that AI commerce is now a major channel, which requires a fundamentally different playbook,” said Alex Sherman, co-founder and CEO of Bluefish. “The brands winning here rewired their holiday strategy around high-quality owned and earned content.”

From “best deals” to “best gifts”

The report also captures a subtle but meaningful narrative shift inside AI systems as the season progressed.

During Black Friday, AI assistants leaned heavily into “best deals” language, surfacing discounts and doorbusters. But as December unfolded, that framing faded fast. Bluefish found that the influence of “best deals” content dropped by more than 30% heading into Christmas.

Taking its place: “best gifts.”

Gift guides—especially those focused on intent-driven queries like “best gifts under $100”—grew steadily more influential as AI assistants shifted from bargain hunting to thoughtful curation. This mirrors how human shoppers behave, but the speed and clarity of the transition inside AI systems caught many marketers off guard.

For brands still optimizing holiday content purely around discounts, the data suggests a missed opportunity.

A few pages shaped most AI answers

Using its proprietary Impact Score and Influence Rank analytics, Bluefish identified another striking pattern: AI recommendations were disproportionately shaped by a relatively small group of high-signal sources.

Publishers including Reddit, CNET, RTINGS.com, PCMag, and lifestyle titles like Who What Wear and Vogue emerged as outsized drivers of AI answers. While some didn’t dominate raw citation counts, their content carried far more weight in how AI systems described, compared, and ranked brands.

In other words, not all citations are created equal. Being mentioned once on the right site often mattered more than being mentioned dozens of times elsewhere—a reality that complicates traditional SEO and PR metrics.

Brands that “won” AI holiday commerce

Within this AI-shaped environment, certain brands consistently surfaced as holiday “winners” by aligning content with the narratives AI prioritized.

In beauty, Ulta stood out for its disciplined “best gift” positioning, reinforced across its own properties and third-party editorial coverage. The result: AI assistants repeatedly surfaced Ulta as a default recommendation.

In luxury, brands like Louis Vuitton, Gucci, and Ralph Lauren benefited from something harder to manufacture quickly—decades of cultural relevance. Dense coverage across curated gift guides and editorial lists positioned them as near-automatic answers to queries like “best luxury gifts,” even when newer competitors were aggressively marketing.

What this means for 2026 planning

The report suggests that leading marketing teams are already adjusting. Instead of treating AI as a black box, they’re:

• Measuring AI visibility on a weekly basis

• Treating AI commerce as a distinct performance channel

• Prioritizing fewer, higher-impact content placements over broad coverage

Looking ahead, Bluefish expects direct AI advertising to begin formalizing in 2026, adding yet another layer of complexity. But the company argues that ads alone won’t solve the core challenge: understanding how AI represents a brand and which sources actually shape that representation.

To that end, Bluefish says it is evolving its platform to help brands identify true sources of AI influence and take systematic action—before AI assistants become the default shopping interface for consumers.

For marketers, the message is blunt: if your brand isn’t legible, trusted, and well-positioned in AI’s world, it doesn’t matter how loud your campaigns are elsewhere.

Get in touch with our MarTech Experts.

">