What is Agentic Commerce for Merchants?
Agentic commerce moves the buy decision to an AI agent that shortlists before the shopper clicks. Merchants need to be visible, citable and convertible to AI agents.
Monday morning. You open ChatGPT and ask for the three best brands in your category. Three competitors come back, with reasons, and your store is missing. This piece is the to-do list for that Monday.
If you sell things online, your storefront is no longer the first surface a shopper sees. In 2026, a shopper has often talked to an AI agent before clicking through to a product page. That agent is ChatGPT, Claude, Perplexity or Gemini. Or it is a vertical agent inside a retailer's own app. That conversation made a recommendation, narrowed a shortlist, often picked a single SKU. The PDP gets the last 30 seconds. The agent had the first 10 minutes.
In this article
That shift has a name: agentic commerce. Distribution has not moved this much since marketplace ranking became the whole job in 2014. The buying funnel is being intermediated by software that reads, summarises, compares, and decides. If that software cannot read your store, you are out of the funnel.
What "agentic commerce" actually means
The phrase is doing a lot of work in trade press, and most of it is wrong. Agentic commerce is not a chatbot bolted onto a checkout page. It is not a Shopify app that summarises reviews. It is the shape of a transaction where an autonomous agent, not a human clicking a search result, makes the whole discover-evaluate-decide pass on the shopper's behalf.
Three things must be true for a transaction to be agentic:
- A software agent received a goal in natural language ("find a linen jacket under £180 that does not wrinkle").
- The agent retrieved candidates from multiple sources (search index, retailer catalogues, review platforms, structured PDPs).
- The agent narrowed the candidates and either presented a shortlist or completed the purchase end-to-end with delegated payment.
The third condition is where the merchant economics flip. When the agent is also the buyer, your job as a brand stops being "rank above competitor X on Google" and becomes "get selected from the candidate set the agent already assembled". The selection criteria are mostly invisible to you. They are inferred from your structured data, your verified-buyer signals, and your resource hub.
The three eras of online distribution
To make sense of agentic commerce, look at what came before.
- 1Era 1, Search-engine optimisation (≈ 2003–2018). Goal: rank on the keyword. Tactic: links, content, structured data, page speed.
- 2Era 2, Social commerce + paid acquisition (≈ 2014–2024). Goal: get clicked on. Tactic: creative volume, audience targeting, attribution.
- 3Era 3, Agentic commerce (≈ 2024–2030). Goal: get cited and selected. Tactic: machine-readable evidence, verified trust signals, agent-friendly catalogues.
Each era added a layer; it did not replace the previous one. PDPs still need to rank in Google. Stores still need paid social campaigns to seed demand. But the return on effort moves. In 2014, the marginal dollar went to bids. In 2026, the marginal dollar goes to making your evidence (your reviews, your guides, your product attributes) accessible to agents.
The merchant checklist
If you accept that agentic commerce is the third distribution era, the work falls into six buckets. Most teams have parts of bucket one and almost none of two through six.
1. Verified-buyer evidence
Reviews, photos, videos and Q&A, tied to specific SKUs, with a cryptographic chain back to a real verified purchase. Agents weight verified evidence orders of magnitude higher than brand copy. If your reviews are unverified Trustpilot-style submissions, agents will skip them when assembling shortlists.
2. Structured product data
Product, Review, AggregateRating, Q&A, BreadcrumbList, FAQ, all the standard JSON-LD shapes, validated and refreshed every time the underlying data changes. This is the substrate every agent reads first.
3. Agent-friendly content
Your resource hub (guides, FAQs, comparison pages, sizing notes) should be written as "question → two-sentence answer → supporting evidence". This is the shape agents quote verbatim. Brands that publish in this format consistently get cited inside three weeks; brands that publish in 1,200-word SEO essays get skipped.
4. Open AI-bot policy
GPTBot, ClaudeBot, PerplexityBot, Google-Extended must be in your robots.txt with explicit allow rules. The number-one reason brands report dropping out of AI answers is a blanket disallow that was set when AI crawlers first appeared and never revisited.
5. Catalogue feeds for agents
Product feeds matter beyond Google Shopping. Agents pull from JSON product feeds, BigCommerce APIs, Shopify Storefront APIs, and increasingly from llms.txt-style indexes. Keep them clean, fresh, and per-SKU complete (size, weight, materials, fit, country of manufacture).
6. Citation analytics
You cannot improve what you cannot measure. Set up a citation dashboard that tracks: which AI engines mention your domain, which URLs they cite, which queries trigger the citation, and how those translate into referral sessions and checkouts.
The six buckets of agent-ready work
- 01
Verified-buyer evidence
Reviews, photos, videos and Q&A tied to SKUs and real purchases.
- 02
Structured product data
Product, Review, AggregateRating, FAQ and other JSON-LD, kept fresh.
- 03
Agent-friendly content
Question, two-sentence answer, then the evidence.
- 04
Open AI-bot policy
GPTBot, ClaudeBot, PerplexityBot and Google-Extended allowed in robots.txt.
- 05
Catalogue feeds for agents
Clean, fresh feeds with every attribute per SKU.
- 06
Citation analytics
Which engines cite you, which URLs, which queries, which sales.
What it actually changes for the merchant
This is not an abstract trend you can watch from the sidelines. Three line-items in your operating budget move once agents are doing the discovery work.
- Acquisition cost: lower at the top of funnel (agents do free recommendation work for visible brands), higher at the bottom (agents are demanding on evidence and will deprioritise weak claims).
- Content production: shifts from blog volume to evidence depth. One verified buyer photo + 12 attribute rows on a PDP > four 2,000-word SEO essays.
- Tech stack: review platform, UGC platform, structured-data platform converge. The brands that win run a single source-of-truth across all three surfaces.
How Idukki fits
Idukki is the layer between your storefront and the agents reading it. We collect verified-buyer reviews, UGC and Q&A; bind them to SKUs with the right JSON-LD; publish a clean llms.txt; and run a citation dashboard that tells you which agents quote which pages. The same infrastructure that drives your PDP gallery drives your agent-visibility, one platform, one data model.
If you want to see the playbook end-to-end, our Answer Engine Optimisation page lays out the full five-step blueprint. The UGC statistics 2026 page has the empirical benchmarks behind the claims in this article.
Closing
Agentic commerce will not be the only way people buy in 2030. Humans will still browse Instagram and tap PDPs forever. But it becomes the layer that gates the rest: if the agent does not pick you, the human never gets the chance to. Treat that as a design constraint from day one.
41%
of 18-34s use GenAI to search products
Bazaarvoice 2025 SEI
$2.6T
Global social commerce 2025
eMarketer
+108%
TikTok Shop US growth YoY 2025
eMarketer
Sources & notes
- 1Bazaarvoice, 2025 Shopper Experience Index · +144% conversion / +162% RPV among UGC-engagers; +354% conversion on PDPs with reviews vs without.
- 2Model Context Protocol (Anthropic) · Open standard for exposing tools that AI agents can invoke. Transport layer of agentic commerce.
- 3Schema.org · Shared vocabulary for structured-data markup. Defines Product, Review, AggregateRating, FAQPage, DefinedTerm types.
- 4Google web.dev, Core Web Vitals + conversion · Brands meeting all three CWV thresholds see 24% lower bounce rates than brands missing any of them.
Continue reading
5 pieces in this clusterThese long-form pieces on the Idukki blog link back to this article, go deeper on the cluster.
- Agentic commerce
Salesforce's Agentforce Commerce just went GA. Here's what it means if you're not on Salesforce.
Shopper Agent, Buyer Agent and Merchant Agent are now generally available, with ChatGPT and Google Search integration coming. What the release actually ships, and what it signals for merchants on Shopify, BigCommerce, WooCommerce or Wix.
- Agentic commerce
Brand Authority for Agents: When a $0 Backlink Beats a $50k Campaign
AI agents grade brand authority differently than Google: one Wikipedia entry, honest Reddit thread or trade-press citation can outweigh a six-figure PR campaign.
- AI search
Answer Engine Optimisation for Ecommerce: The Complete AEO Library
Being cited by ChatGPT, Perplexity and Google AI Overviews is the new ranking. The full AEO library for ecommerce: crawlability, schema, evidence and measurement.
- Agentic commerce
MCP, agent tool calls and product feeds: the merchant’s guide to what to implement first
Shopify made catalogue, cart and checkout tool-calling free and automatic in August 2026. Here is what that changes about MCP, OpenAI tools and Google’s agent feed, what it deliberately does not give an agent, and a vendor-ready RFP template.
- Strategy
Grounded answers plus a human handoff beat fully autonomous AI on the PDP
Full autonomy is the wrong default on a PDP, where a confident wrong answer costs a return and a one-star review. Grounded answers plus a clean handoff convert better.
More from Rohin Aggarwal
- Agentic commerce
The Agentic Commerce Stack: a reference architecture for merchants in 2026
Seven layers sit between a shopper and your checkout. A reference architecture plus a 90-day plan for being visible, citable and convertible inside AI shopping agents.
- Agentic commerce
Will AI agents recommend your store? A 2026 readiness read
An agent recommends you only if it can reach, parse, trust and quote your catalogue. If it cannot, you never make the shortlist. Here is how to tell where you stand.
- Agentic commerce
The 7-second window: why agentic commerce makes your PDP the new email subject line
Shopping agents quote roughly 700 words of your PDP. The rest of the page goes unread. Whatever they quote wins the sale, so the top of the page has to carry the answer.