By Sam Hall
How to identify and verify AI agent requests in your logs
Find AI crawler and agent requests in your logs, check provider identity, and keep verified, claimed and unknown evidence distinct.
Insights & Field Notes.
Written analysis of merchant-observable agent traffic — what request logs actually show, what they cannot show, and how to tell the difference.
- By Sam Hall
AI referral traffic and agent activity: what each tells you
Separate AI referrals, verified agent requests and recorded shopping actions. A practical reporting guide for merchant analytics and engineering teams.
- By Sam Hall
How to diagnose a gap between an AI agent visit and checkout
Find the last recorded cart or checkout milestone, distinguish errors from handoffs, and document evidence gaps without assuming a lost sale.
- By Sam Hall
Your best price may be invisible to an AI shopper
Before issuing another coupon, check whether an AI-assisted shopper can use the benefits they already have. A practical loyalty test for merchants.
- By Sam Hall
A cart can transfer perfectly and still fail the customer
The right product can reach checkout under the wrong terms. Our view on Google’s UCP updates, and the exception tests merchants should run before scaling.
- By Sam Hall
Would you recognize an AI shopper?
What merchants can actually learn from AI agent traffic, where to look for the evidence, and why a missing checkout is not automatically a lost sale.
- By Sam Hall
The AI Shopper Is Here: Retail Is Still Catching Up
AI shopping is changing product discovery, retail traffic and product data. What the research shows, and why merchants still struggle to see the full journey.
What this is
A small, slow series. Each entry carries a named author, a visible date — a publication date for published entries, a preparation date for drafts — and a visible source list, and every factual claim in it is attributable to something a reader can check.
Entries describe merchant-observable evidence only: request-level signals recorded on surfaces the merchant controls. Cartograph does not claim to know what an external AI model concluded, preferred, recommended, or decided off-property, and absence of agent evidence does not prove human traffic.
What you will not find here
- No benchmark numbers or traffic volumes without a linked source.
- No inferences about model behaviour observed outside merchant surfaces.
- No customer data, and no aggregate figures drawn from private deployments.
- No placeholder or preview entries standing in for work that does not exist.
Explore the evidence framework
Explore the evidence framework and practical guides behind these field notes.
- How Cartograph is designed to workReadiness framework, evidence labels, and explicit scope boundaries.
- Market evidenceSourced demand-side evidence, and what it does not establish on its own.
- Agent commerce evidenceRequest-level signals AI agents leave behind on merchant surfaces.
- What is agentic commerceDefinition for merchants, and which parts are observable on your own property.
- AI shopping agentsHow shopping agents read product data and what those requests leave in logs.
- AI agent readiness checklistAudit across product data, storefront, checkout, logging, and crawler policy.
- Provider-documented AI crawlers and agentsSource-backed reference for crawlers, fetchers, and identity verification.