How Purchase Decisions Now Form Before the Customer Is Involved

A growing share of product research, comparison, and purchasing now runs through an AI agent rather than a person. The agent interprets the request, narrows the field, and returns a small set of purchase-ready options, and the customer chooses from those. This is measurable.

Adobe Analytics recorded generative-AI-referred traffic to US retail sites growing roughly 690% year over year across the 2025 holiday season (Adobe Analytics, 2026). Capgemini's consumer research finds that 58% of consumers have replaced traditional search engines with generative-AI tools as their primary route to product recommendations (Capgemini Research Institute, 2025). McKinsey estimates conversion from AI-generated product recommendations at about 4.4 times that of traditional search (McKinsey, cited in MetaRouter, 2026).

The funnel still has the same stages. They have been compressed into one interaction that the customer never observes. By the time a person sees options, the consideration set has already been built, and whatever sits outside it is, for that purchase, absent. This is the part most brands have not yet absorbed: the decisive moment has moved upstream of every surface they currently optimize.

What I propose is a Brand Visibility for Agentic Commerce (BVAC) framework and diagnostic that enables brands to create more opportunity and thrive in an agent-to-agent world.

The agent reads a different document than the customer

A human shopper responds to narrative, design, and merchandising. An agent assembling a shortlist reads structured product data — identifiers, attributes, policy fields, ratings, certifications — and treats unstructured marketing copy as low-signal. The investment that wins a person's attention is largely invisible to the system that now decides whether a person sees the brand at all.

The behavior driving the exclusion is risk aversion. An agent that recommends a product it cannot stand behind incurs the costlier error, so when an expected attribute is missing or ambiguous, it skips the product rather than guessing. The exclusion is not a judgment that the brand is worse than the alternative. It is a judgment that the brand cannot be assessed, and from outside the agent, those two outcomes are indistinguishable. Stibo Systems calls the result decision invisibility: the brand is filtered out before a human is involved, so there is no bounce-rate change and no campaign to diagnose (Molino Sánchez, 2026).

The evidence on what agents actually weigh

The clearest read on agent behavior comes from controlled testing rather than inference. Sabbah and Acar evaluated eight standard promotional mechanisms across four models and more than 16,000 simulated consumer purchase decisions for a 2026 Harvard Business Review study, using a deliberately ordinary product spread. One signal, structured ratings, moved selection upward consistently across every model and product category. Strike-through pricing, countdown timers, and bundling showed no stable pattern, and bundling reduced selection in at least one case. The reasoning models were the more skeptical, in several cases appearing to penalize overt persuasion as a quality signal in itself (Sabbah & Acar, 2026).

The finding is consumer-scoped and does not extend to every category, but its implication is general. A promotional layer that many merchandising teams treat as costless can suppress selection on the surface that an agent reads. The remedy for being skipped is structural, and several of the tactics built to persuade a person work against the brand when an agent is the reader.

Why does the loss not appear in the dashboard

Decision invisibility persists because it is inherently invisible. Discovery and comparison happen inside the agent; the customer, when they arrive, arrives later through a direct or organic path, and the influence that determined the decision leaves no trace in the merchant's analytics (Hanna, 2026). Query fan-out and zero-click shortlisting mean the brand sees neither the queries it lost nor the comparisons it was excluded from.

This creates a specific hazard in how the problem is prioritized. Every other category of marketing problem announces itself through a metric that moves, and the triage process that allocates attention to the dashboards that change will pass over decision invisibility, because the dashboards do not change. The financial logic compounds it: a loss that cannot be attributed does not enter the return-on-investment calculation, and a cost that cannot be priced is reliably under-prioritized against costs that can. The absence of a signal is the expected presentation of the problem, not evidence that the problem is small.

A concrete version of the failure

Consider a brand with a strong premium position in its category. Its flagship product justifies its price through materials, construction, and warranty terms, all of which the brand communicates persuasively in copy and imagery written for a human reader. None of it is encoded as structured, verifiable attributes. An agent comparing products in the category works from the intersection of the structured fields it can read, finds the premium product described by the same default attributes as cheaper entries, and ranks it on price against products it was never meant to compete with. The brand's positioning is intact, and its data surface erases it. The more frequent and more expensive error in practice follows the same pattern: investing in a strategic capability. At the same time, a prerequisite goes unaddressed, such as a quarter spent standing up an agent initiative while the return policy remains unmarked in unstructured HTML.

What this changes

The inputs that set the urgency — agent adoption, and the data quality of competitors and marketplaces — sit outside any single brand's control and are moving in one direction. The variable a brand does control is its own legibility to the systems now making the selection, and it is the variable most often deferred, because nothing in the current measurement stack reports its decline. Knowing whether an agent can resolve a brand, compare it on equal terms, and find structured grounds to trust it is now a question worth answering deliberately rather than leaving to be answered by omission. A diagnostic discipline such as the Brand Visibility for Agentic Commerce Framework exists to make that question answerable; the prior step, and the point of this piece, is recognizing that the question is now the one that determines whether the brand is in the set at all.

Learn more how I can help here: Consulting | Workshop

References

Adobe Analytics. (2026). AI-driven traffic surges across industries, retail sees biggest gains [2025 holiday shopping recap]. Adobe.

Capgemini Research Institute. (2025). What matters to today's consumer (4th ed.). Capgemini.

Hanna, C. (2026). Attribution gap in agentic search: How to close it. Semrush.

MetaRouter. (2026). Agentic commerce trends and statistics for 2026. MetaRouter. (Conversion figure attributed to McKinsey.)

Molino Sánchez, M. (2026). 7 signs your brand is losing ground in agentic commerce. Stibo Systems.

Sabbah, J., & Acar, O. A. (2026, May 12). Research: Traditional marketing doesn't work on AI shopping agents. Harvard Business Review.