The model Spoke 02 · Future-proofing

Agentic Commerce

Agentic commerce is shopping through AI assistants and agents such as ChatGPT, Gemini, Perplexity or Claude. The shopper states a goal in plain words. The agent works out what is needed, compares the options and, more and more often, books and buys on the shopper's behalf.

See this week's updates
One intent, one solutionA chat bubble says: I'm a wedding guest in 3 weeks and want my hair to look great. An AI agent plans the whole solution bundle: a hairdresser booked for Saturday at 10:00 (a service), an at-home routine of three products, delivery that arrives in 2 days, and reminders. Everything leads to Wedding-ready. Before: 4 separate searches. Now: 1 intent."I'm a wedding guest in 3 weeksand want my hair to look great."AI agentplans the whole solutionSolution bundleServiceHairdresser bookedSat 10:00ProductsAt-home routinemask, protectant, stylingDeliveryArrives in 2 daysin time for the dayRemindersSet and readywhen to use whatWedding-readyBefore: 4 separate searchesNow: 1 intentOne intent, one solution"I'm a wedding guest in 3 weeksand want my hair to look great."AI agentplans the whole solutionSolution bundleServiceHairdresser bookedSat 10:00ProductsAt-home routinemask, protectant, stylingDeliveryArrives in 2 daysin time for the dayRemindersSet and readywhen to use whatWedding-readyBefore: 4 separate searchesNow: 1 intent
Simplified view of the direct commerce model.

Why it matters for brands

Shopping is moving from keywords to intent.

Today: one search per product. "Hair mask." "Heat protectant." "Hairdresser near me." The shopper compares, books and orders, and stitches it all together alone.

With an agent: one goal. "I'm a guest at a wedding in 3 weeks and want my hair to look great." The agent builds the whole solution, services and products together:

  • Book a hairdresser appointment that fits the calendar.
  • Pick an at-home routine: a repair mask, a heat protectant, a styling product.
  • Order it to arrive in time.
  • Remind the shopper when to use what.

Then it checks out and books, within limits the shopper sets.

Ranking for one keyword is no longer enough. A brand has to be part of the solution bundle, and agents have to be able to read it and book it:

  • Structured product data that says what a product does and for whom.
  • Availability of stock, prices and appointment slots.
  • APIs and protocols that let an agent check out or book.
  • Trust signals such as reviews, clear policies and secure payment.

How it connects to the hub

The agent handles the moment. The hub keeps the relationship.

The D2C hub owns what the agent needs: the product data, the routine and the customer relationship. It knows which products belong together, when the routine runs out and what the customer bought last time. That is why the agent comes back to the brand the next time the shopper asks.

Key facts

This 2025 holiday season, consumers embraced generative AI more than ever as a shopping assistant in their purchasing decisions.

Vivek PandyaLead analyst, Adobe Digital Insights (January 2026)Source
Weekly

See this week's updates

7 updates and this week's reading list on Agentic Commerce, every item linked to its source.

See this week's updates

Explore the model