AI agents are becoming a new layer between shoppers and brands. WARC and PHD project agent-facilitated spending will rise from $944 billion in 2026 to $3.35 trillion by 2030, yet BCG finds only 4% of consumers would let an agent complete a purchase alone. The fight today is for the recommendation, not the checkout.
Quick answer
Agentic commerce is shopping where an AI agent researches, compares and sometimes buys on a person's behalf. To prepare, make your product data accurate and machine-readable, keep price and stock current, and earn trusted third-party signals such as reviews. Do this before Black Friday on 27 November, because agents can only recommend what they can read and verify.
In this guide: what it is · the numbers · which sectors move first · how agents choose · a 7-step checklist · FAQ
What agentic commerce is
Agentic commerce is shopping in which an AI agent handles discovery, comparison and sometimes checkout for the shopper. A chatbot answers questions and a search engine returns links, while an agent works toward a goal. BCG's example: a shopper asks an assistant to order household basics, find a gift under $50 and check a shoe price drop, and the agent handles all three without the shopper visiting a page or seeing an ad. In that flow, a retailer "can make a sale and never receive a visit", as BCG puts it.
WARC and PHD describe agents as a new decision-making layer between consumers and businesses. Their "Four modes" framework says four kinds of influence will coexist: brand to consumer, consumer to consumer, agent to consumer and agent to agent. Which mode dominates depends on the category and the purchase occasion.
What the numbers say
Adoption of AI for research is far ahead of delegation of purchases. Three recent studies show both sides.
| Source | Sample / date | Finding |
|---|---|---|
| WARC and PHD (22 Jul 2026) | Global forecast | Agent-facilitated consumer spending rises from $944B in 2026 to $3.35T in 2030, or 3.8% of global consumer spending, up from 1.3%. |
| BCG (21 Oct 2026) | 50 top global retailers and 100 top brands | 68% of retailers expect agentic commerce to have a significant or transformational effect within two years. Only 4% of consumers would let an agent finish a purchase alone. |
| BCG, European data | Leading European brands and retailers | Referrals from external agents were under 1% of website visits in late 2025 and could reach 5% by end 2026. |
| Northwestern Retail Analytics Council and Minty (Sep 2026) | 150 senior commerce decision-makers | 79% expect their main customers to use AI tools in 2027. Only 18% expect traditional search to be their main route to shoppers. |
Read the numbers with care. WARC and PHD's figures are a forecast. The Northwestern survey records what marketers expect, not what shoppers do, and Minty, its partner, sells an AI shopping assistant. BCG notes that referral counts understate agents' influence, because shoppers who choose inside an AI platform and then visit directly are not counted; including them, the effect on traffic could be up to four times the referral figure.
The gap that matters. BCG says 81% of US consumers expect to use AI agents when shopping, yet only 4% would let an agent buy on its own. Shoppers hand over discovery and comparison first, and checkout last. The recommendation is where brands win or lose today.
Which sectors move first
Agents take over first where choices are frequent, comparison-heavy and data-rich. WARC and PHD name three sectors most likely to shift towards agent-to-agent interactions.
| Sector | Agent-facilitated spending | Why it suits agents |
|---|---|---|
| Telecoms and utilities | $57.6B in 2026 to $410.3B in 2030 (over 600% growth) | Frequent billing cycles and comparison-heavy buying |
| Financial services | Up 235.3% to $237.9B by 2030 | Agents narrow mortgage, insurance and banking options before a person decides |
| Travel and transport | $78.1B to $275.6B | Itinerary planning, price comparison and booking |
The report also expects fast growth in media and publishing, food, retail and soft drinks. Within retail, BCG says the risk is greatest in low-consideration categories such as household consumables and repeat orders, which are easiest to delegate. Fashion, beauty and luxury are less exposed because discovery plays a larger role.
If you sell or publish in a comparison-driven category, your pricing, plan details and terms need to be clear enough for a machine to compare, not just persuasive to a person.
How agents choose
Agents recommend what they can read, verify and compare. BCG says agents evaluate retailers on product data, prices, availability, delivery, returns and offers they can interpret. WARC and PHD add that trust signals may tip the choice.
| Signal | What the agent checks | What you publish |
|---|---|---|
| Product data | Attributes, specs, categories | Complete, structured, consistent product information |
| Price and stock | Whether the offer is current | Accurate live price and availability |
| Delivery and returns | Cost, speed, terms | Clear shipping and returns rules |
| Offers and loyalty | Member value it can calculate | Member prices and benefits stated in comparable terms |
| Trust | Reputation across sources | Reviews, transparent pricing, service quality |
The data is moving from crawled pages to live feeds. BCG reports that major AI platforms are pursuing direct integrations so agents can retrieve live price, availability, delivery and returns information. Among surveyed retailers, only 18% consider their marketplaces ready for agents, while 67% plan to enrich product data and 58% to offer real-time availability within two years.
Share product data, protect customer data. BCG recommends making product, inventory, pricing, offer and fulfilment data accessible to agents while keeping customer identities, loyalty histories and behavioural data under your control. Its survey found 42% of retailers open their data to any agent, 32% give access only to chosen agents, 6% block agents, and 20% are undecided.
Example from BCG's report. Nordstrom cleaned and enriched its product information, fixed technical problems that kept its catalogue out of agent results, and integrated its product feeds with OpenAI, Google and others. It reports strong early results on agent visibility.
A 7-step readiness checklist
Black Friday falls on 27 November and Cyber Monday on 30 November. Work through these in order; the first three matter most before then.
- Audit your product data. Check that every product has a clear title, full attributes, correct category, price, stock status and shipping details, and that your feed matches your site.
- Decide your agent policy. Choose deliberately between open to any agent, selective access, or blocked, then check that robots.txt and your CDN settings match that choice.
- Make price, stock, delivery and returns live and accurate. An agent that finds a stale price or a wrong stock status can drop you from its recommendation.
- State offers in comparable terms. Write member prices, delivery windows and points value so a machine can calculate them. Keep exclusive perks for shoppers who come to you directly.
- Strengthen trust signals. Show genuine reviews, transparent pricing and clear service terms. Never fabricate reviews: agents cross-check sources, and it is deceptive.
- Turn agent-referred buyers into known customers. Use registration and loyalty sign-up to build consented relationships, as BCG advises, and keep investing in distinctive brand storytelling for people.
- Measure and budget for the cost. Test your key shopping questions monthly in ChatGPT, Gemini, Copilot and Perplexity and log whether you are recommended. BCG warns that agent-routed demand is unlikely to stay free, as comparison sites, paid search and travel agencies took a fee, so model it in your margins.
Common mistakes and expert tips
The costliest mistake is treating agents as a future problem while your data is already being read, or ignored, today.
Mistakes to avoid
- Measuring agents by referral clicks alone. BCG says shoppers who choose inside an AI platform and then visit directly are not counted, so referral reports understate influence.
- Blocking agents without a plan. Retailers with strong direct appeal can restrict access, but those that rely on agents for reach may need to share more.
- Sharing customer data along with product data. BCG advises keeping identities, loyalty histories and behaviour under your control. Its survey shows 87% of retailers never share customer data with agents.
- Relying on emotion alone. WARC and PHD say storytelling and distinctive brand assets still matter, but products and brand signals must also be understandable to AI systems.
- Assuming it is only a retail issue. Telecoms, financial services and travel are forecast to be among the biggest agent-led categories.
Tips that pay off
- Stage your rollout. Nordstrom committed first to visibility on external AI platforms and AI-assisted discovery on its own site, and held back on agent checkout until demand justified it.
- Hold third-party sellers to the same data standard. BCG notes Nordstrom does this on its marketplace.
- Track share of recommendation for your top 10 shopping questions, not just rankings.
- Keep reviews and policies in plain text on your pages so both people and machines can read them.
FAQ
What is agentic commerce?
It is online shopping in which an AI agent researches, compares and sometimes completes purchases for a shopper. It differs from a chatbot because the agent works toward a goal on the shopper's behalf.
Will AI agents replace human shoppers?
Not soon. BCG found only 4% of consumers would let an agent finish a purchase alone, and WARC and PHD expect consumers to keep making most purchase decisions over the next few years.
How big is agentic commerce?
WARC and PHD forecast agent-facilitated consumer spending of $944 billion in 2026, rising to $3.35 trillion by 2030. That is a forecast, not a measured total.
How do I make my products visible to AI agents?
Keep product data complete and structured, price and stock current, and delivery and returns terms clear. Build genuine reviews, and consider feeds or integrations offered by the AI platforms you want to reach.
Do small businesses need to act now?
Yes, but in small steps. Cleaning product data and keeping price and stock accurate help with agents and with ordinary search and customers, so the work pays off either way.
Conclusion
Shoppers are handing research and comparison to AI long before they hand over the checkout. That makes the agent's recommendation the new shelf position. The brands that win will keep their product data accurate and readable, keep price and stock live, earn real trust signals, and still give people a reason to buy from them directly. Start with your product feed this week, then test how the major assistants describe your top products.
Want your brand or expertise featured? Blogiefy publishes guest posts and sponsored articles with clear disclosure. [Add link to your submission page] to send a pitch.
Sources
All opened and checked on 10 Nov 2026.
- BCG: How Retailers Can Reinvent the Digital Flywheel for Agentic Commerce (21 Oct 2026; survey of 50 top global retailers and 100 top brands)
- Marketing-Interactive: Global consumer spending via AI agents to hit US$3.35tr by 2030 (22 Jul 2026; reports the WARC and PHD study)
- Digital Commerce 360: How marketing budgets are adapting to AI-driven purchases (17 Sep 2026; reports the Northwestern Retail Analytics Council and Minty survey)
