https://cartforagents.com/Cart for AgentsWhat AI shopping agents know about your products https://cartforagents.com/guidesGuidesPractical guides to product evidence for AI shopping agents. https://cartforagents.com/researchResearchDated, sourced and never paid for. https://cartforagents.com/for-agentsFor agentsHow to read and cite Cart for Agents. https://cartforagents.com/guides/aco-vs-geoACO vs GEO: cited pages and product records are different winsCompare citation, product, variant and offer outcomes with a paired measurement matrix; acronym definitions belong in the SEO/AEO/GEO/ACO comparison. https://cartforagents.com/guides/agent-commerce-optimizationAgentic commerce optimization (ACO): test product evidence before chasing rankA five-stage ACO test plan for one product: confirm eligibility, read the record agents receive, check each shopper requirement against a sourced fact, test one truthful change, then verify the offer. With stop rules, outcome measures and prior art. https://cartforagents.com/guides/agent-friendly-ecommerce-apiAgent-friendly ecommerce: when you need an API and when the page is enoughDiagnose the failed retrieval route, compare faithful outputs and use a decision tree before adding an API or machine-readable endpoint. https://cartforagents.com/guides/agent-ready-product-dataAgent-ready product data: build a factual record for each SKURecord product identity, values, units, test conditions, sources, exclusions and unknowns for each variant. Includes a fill-in template, a worked humidifier record and a page-versus-catalog check. https://cartforagents.com/guides/agentic-commerceAgentic commerce: what Shopify merchants can controlFollow one product through the five stages of agentic commerce, from catalog intake to checkout: who decides each one, what a Shopify merchant controls, and what to check first. https://cartforagents.com/guides/agentic-commerce-protocolAgentic Commerce Protocol: separate product discovery from checkout setupUse an integration checklist to distinguish feed admission, discovery data, checkout contracts and supported purchase authorization. https://cartforagents.com/guides/agentic-commerce-strategyAgentic commerce strategy: estimate effort and break-even before you startUse editable assumptions for fix costs, order contribution and qualified traffic to decide whether to run a bounded merchant test. https://cartforagents.com/guides/ai-product-descriptionsAI product descriptions: open with type, use case, key numbers and limitsWrite the first sentence of a product description for shoppers and AI agents: product type, use case, verified measurements and honest limits, with two worked rewrites. https://cartforagents.com/guides/ai-product-discoveryAI product discovery to purchase: find where the sale stopsFollow one product from recommendation through seller availability, shipping, authorization and order confirmation using a handoff worksheet. https://cartforagents.com/guides/ai-product-visibilityAI product visibility: design a repeatable monitoring planChoose 3 to 5 products, real customer questions, buyer contexts, routes and a recheck schedule; log every check in a 15-column ledger; and keep "returned," "recommended," "facts correct" and "sold" as separate outcomes. https://cartforagents.com/guides/ai-product-visibility-dashboardAI product visibility dashboard: turn each red cell into an owned fixSpecify separate visibility, factual-fidelity and commercial fields with evidence links, action owners and next-check dates; no composite readiness score. https://cartforagents.com/guides/ai-shopping-agentAI shopping agents: compare capabilities and merchant controlsCompare research, product comparison and purchase capabilities with merchant controls over facts, distribution and checkout eligibility. https://cartforagents.com/guides/ai-shopping-for-small-businessAI shopping for small businesses: where a small DTC brand can winCheck eligibility, buyer constraints, evidence gaps and operating costs before choosing a product cohort or assuming bigger brands always win. https://cartforagents.com/guides/ai-shopping-personalizationAI shopping personalization: control context before comparing resultsCompare thin and richly constrained buyer profiles, record memory and country settings, and preserve product disqualifiers in a context-control card. https://cartforagents.com/guides/ai-shopping-platformsAI shopping platforms: compare data routes and checkoutCompare ChatGPT, Google AI Mode and Gemini, Perplexity and Microsoft Copilot by how your product data reaches each one, what changes results, and where checkout happens. https://cartforagents.com/guides/ai-shopping-prompt-trackingAI shopping prompt tracking: build test prompts from real customer questionsBuild a CSV panel with shopper situations, hard constraints, preferences and evidence sources; Google keyword volume is not agent-prompt demand. https://cartforagents.com/guides/ai-visibilityAI visibility: distinguish mentions, citations and product outcomesClassify a brand mention, source citation, product retrieval, recommendation and completed order with examples of what each does not prove. https://cartforagents.com/guides/ai-visibility-auditProduct-level AI visibility audit: find the next factual fixRun a one-product audit with shopper constraints, received evidence, unknowns and an owner-assigned fix worksheet, not a readiness grade. https://cartforagents.com/guides/ai-visibility-checker-accuracyAI visibility checker accuracy: test what each checker actually readsTest an AI visibility checker before you trust it: what it fetched, which fields it read, how it handles missing data, and a rubric for judging any product-URL check. https://cartforagents.com/guides/ai-visibility-scoreAI visibility score: check the denominator before you trust the numberAsk how a score samples products, contexts and time, handles missing data and weights components using an independent score-review checklist. https://cartforagents.com/guides/amazon-rufus-optimizationAmazon Rufus optimization for DTC brands: find the Amazon route your products are onUse a naming timeline and channel map to distinguish marketplace listings, off-Amazon offers and merchant-controlled product evidence. https://cartforagents.com/guides/chatgpt-product-feedChatGPT product feeds: map fields, approval and schema conflictsMap source fields to the documented feed contract, preserve variant identity and resolve conflicting examples without mixing schema versions. https://cartforagents.com/guides/chatgpt-visibility-trackingChatGPT visibility tracking: a repeatable protocol for manual checksRecord market, account context, model versus product, returned representations and repeated outcomes using a fixed protocol and failure log. https://cartforagents.com/guides/copilot-shoppingCopilot shopping: check merchant data, engagement and sponsored resultsCompare merchant feeds, web sources, likely engagement and sponsorship statements; record what Microsoft documents and what remains unknown. https://cartforagents.com/guides/generative-engine-optimization-ecommerceGenerative engine optimization for ecommerce: a merchant task backlogPrioritize crawlability, source evidence, product facts and feed consistency using owners, acceptance checks and a worked merchant backlog. https://cartforagents.com/guides/geo-vs-seoSEO, AEO, GEO and ACO: which merchant job does each cover?Match search indexing, answer citation, product evidence and purchase execution to the responsible team and a measurable outcome. https://cartforagents.com/guides/google-ai-mode-shoppingGoogle AI Mode and Gemini shopping: check your merchant data routesSeparate indexed web content, Merchant Center data and checkout eligibility using a route teardown and merchant verification checklist. https://cartforagents.com/guides/google-merchant-center-free-listingsGoogle Merchant Center free listings: check eligibility for Gemini and ShoppingGet Google Merchant Center free listings approved and eligible for Gemini: permission, product status, required attributes and price parity, in the order to check them. https://cartforagents.com/guides/google-ucp-checkoutGoogle UCP checkout: verify eligibility, exclusions and rolloutRun an account-and-market preflight using a dated rollout conflict table, unsupported-product checks and explicit unknowns. https://cartforagents.com/guides/gtin-mpn-skuGTIN, MPN, SKU and UPID: choose the correct product identifierUse an identifier decision table to distinguish manufacturer, merchant and catalog identities, including legitimate missing-ID cases. https://cartforagents.com/guides/how-ai-shopping-agents-choose-productsHow AI shopping agents choose products: what the evidence showsSeparate official selection inputs, controlled experiments and vendor observations in a stage-by-stage ledger with methods and limitations. https://cartforagents.com/guides/how-to-rank-in-chatgptHow to rank in ChatGPT shopping: four checks for product recommendationsCheck admission, received product facts, shopper context and merchant offers in sequence; separate documented inputs from ranking guesses. https://cartforagents.com/guides/humidifier-field-guideHumidifier field guide for shopping agents: questions to ask, evidence to require, when none fitAsk about room size, noise and runtime; require values with units and test conditions; separate unknown from incompatible; say none fit when no humidifier meets a hard requirement. A neutral rubric with red flags and a worked example. No product picks. https://cartforagents.com/guides/humidifier-product-dataHumidifier product data: publish noise, room coverage and runtimeMap noise, area, output, tank and runtime claims to modes, conditions and sources, using a dated numeric-versus-vague Catalog example. https://cartforagents.com/guides/machine-readable-product-dataMachine-readable product data: keep JSON twins matching the visible pageUse a versioned product-evidence JSON example and parity checks for values, scope, sources and unknowns; no special-file ranking promise. https://cartforagents.com/guides/meta-muse-shoppingMeta Muse shopping: distinguish catalog presence from browser discoveryHow Meta's Muse agent finds products through your catalog and through its own browser, and which merchant facts to check on each route. https://cartforagents.com/guides/negative-product-attributesNegative product attributes: publish who and what your product is not forRepresent not-for statements and fit exclusions in a proposed fact schema with merchant examples, not instructions that override a shopper. https://cartforagents.com/guides/perplexity-shoppingPerplexity shopping: check merchant admission and data routesSeparate public-web discovery from merchant-program admission, specification sharing and historical checkout arrangements in a route checklist. https://cartforagents.com/guides/pet-product-dataPet product data: make paw size and weather claims checkableUse paw or body measurements, size conditions and weather-claim sources; separate product variety from seller variety in a dated shelf example. https://cartforagents.com/guides/product-claim-provenanceProduct claim provenance: tie every claim to its value, scope and sourceUse a proposed JSON claim record with variant binding, units, conditions, provenance and unknown states; it is not an industry protocol. https://cartforagents.com/guides/product-constraint-coverageProduct constraint coverage: check evidence for each shopper requirementMap a known shopper brief to supporting, missing or conflicting facts; separate evidence coverage from product quality and recommendation probability. https://cartforagents.com/guides/product-data-freshness-for-aiProduct data freshness for AI: what to recheck and how oftenSet recheck triggers for price, inventory, policies and specifications; distinguish real verification dates from decorative updated-year labels. https://cartforagents.com/guides/product-evidence-graphProduct evidence graphs: keep variant, component and seller facts in placeModel product, variant and offer relationships with explicit inheritance and exclusion edges; keep sibling-choice experiments separate. https://cartforagents.com/guides/product-feed-optimizationProduct feed optimization for AI shopping: map and validate every fieldTrace source facts through transformations to feed output using field mappings, validation rules, variant checks and explicit unknowns. https://cartforagents.com/guides/product-feed-price-mismatchProduct feed price mismatch: reconcile page, feed and structured dataFix price and stock mismatches between your product page, feeds and structured data: pick the source of truth, compare timestamps, and tell lag from a mapping bug. https://cartforagents.com/guides/product-material-dataProduct material data: tie composition and certification claims to proofSeparate composition, performance and certification claims using source, issuer, scope and expiry fields in a substantiation worksheet. https://cartforagents.com/guides/product-not-showing-in-chatgptProduct not showing in ChatGPT? Find the stage that's failingWork through admission, retrieval, facts, market context and offers with an evidence log before changing product copy. https://cartforagents.com/guides/product-page-prompt-injectionProduct-page prompt injection: treat ranking manipulation as a security failureFind and remove product-page text written to manipulate AI agents, including copy you didn't write, with a defensive checklist and what to publish instead. https://cartforagents.com/guides/product-schema-markupProduct schema markup: match every marked-up value to the visible pageAn annotated Product and Offer JSON-LD example, and the checks that keep every marked-up price, stock, identifier and policy value identical to the visible page. https://cartforagents.com/guides/product-specification-templateProduct specification template: comparable values, units and test conditionsUse a specification template to retain units, measurement conditions, source dates and uncertainty instead of treating bare numbers as comparable. https://cartforagents.com/guides/product-variant-structured-dataProduct variant structured data: tie size, colour, price and stock to the right SKUBuild an identity and inheritance matrix that preserves buyer-relevant sibling differences and prevents facts leaking between variants. https://cartforagents.com/guides/share-of-constraint-coverageShare of constraint coverage: calculate it on a real shopper briefCalculate a proposed evidence-coverage measure on a worked humidifier brief, keeping unknown, conflicting and inapplicable facts separate. https://cartforagents.com/guides/shipping-return-policy-structured-dataShipping and return policy structured data: attach each promise to the right offerMap destination, handling, transit, return windows and SKU exceptions to offers, with source dates and unsupported destinations explicit. https://cartforagents.com/guides/shopify-agentic-storefrontsShopify Agentic Storefronts: map merchant and platform responsibilitiesMap Shopify Catalog eligibility, data mapping, variant grouping, generated fields, distribution routes, crawler and opt-out controls, and downstream assistant and checkout decisions to what Shopify, the merchant, the platform and the buyer control. https://cartforagents.com/guides/shopify-agentic-tabShopify Agentic tab: interpret listing quality and search previewsRead the public admin documentation with an annotated walkthrough, separating preview evidence from downstream assistant recommendations. https://cartforagents.com/guides/shopify-catalog-requirementsShopify Catalog requirements: check each gate when a product is excludedCheck every Shopify Catalog gate in order, from store standing to product rules, to find out whether a product was excluded or simply outranked. https://cartforagents.com/guides/shopify-chatgptShopify and ChatGPT: check what is automatic and what you controlSeparate Shopify distribution from merchant-owned product facts, eligibility checks and downstream offers in an ownership checklist. https://cartforagents.com/guides/shopify-global-catalogShopify Global Catalog: distinguish products, UPIDs and seller offersTell search results, UPID clusters, variants and seller offers apart in Shopify's Global Catalog, with an identity diagram and the unit to use for every count. https://cartforagents.com/guides/shopify-metafields-aiShopify metafields for AI: verify which facts reach Shopify CatalogMap merchant-entered fields to returned records, distinguish inferred attributes and leave unobserved mappings unknown. https://cartforagents.com/guides/showing-up-in-ai-shopping-cartsShowing up in AI shopping carts: what merchants can optimize at each stepThe four steps from AI recommendation to completed order, and what a merchant can optimize at each one: facts, variants, price, shipping, policies and checkout settings. https://cartforagents.com/guides/size-chart-accessibilitySize chart accessibility: publish fit measurements as a text tableBuild an accessible measurement table with garment versus body dimensions, units, tolerances and fit caveats alongside any size-chart image. https://cartforagents.com/guides/spec-sold-product-guidesSpec-sold products: turn shopper requirements into category facts and rule-out conditionsTranslate buying requirements into category fields, units, conditions, sources and rule-out conditions, with a worked humidifier worksheet and short contrasts for winter clothing, travel bags and cosmetics. https://cartforagents.com/guides/track-chatgpt-traffic-ga4Track ChatGPT traffic in GA4: set up an AI assistant channelSet up GA4's AI Assistant channel or a stricter custom channel group for ChatGPT, Gemini, Copilot and other assistants, with a tested regex and a reading checklist. https://cartforagents.com/guides/travel-product-dataTravel product data: carry-on dimensions, capacity and airline fitPublish carry-on claims shoppers and agents can check: external dimensions, loaded state, capacity method, and the airline limits you compared against, with dates. https://cartforagents.com/guides/ucp-vs-acpUCP vs ACP: compare contracts before choosing an integrationCompare protocol roles, discovery and execution boundaries with version notes, supported capabilities and unresolved interoperability questions. https://cartforagents.com/guides/unit-pricing-ecommerceUnit pricing for ecommerce: compare multipacks and bundles fairlyRecord pack counts, net contents, currency and variant scope in a unit-price worksheet; keep bundle checkout eligibility separate. https://cartforagents.com/guides/universal-commerce-protocolUniversal Commerce Protocol (UCP): capabilities, checkout and merchant controlsWhat the Universal Commerce Protocol covers for merchants: the /.well-known/ucp profile, catalog, cart, checkout and orders, read against a real store profile. https://cartforagents.com/guides/winter-clothing-product-dataWinter clothing product data: temperature ratings, scope and fitPublish temperature units, activity and layering conditions, garment fit and rating provenance; use our corrected matcher example as a warning. https://cartforagents.com/research/agent-shopping-researchAI shopping product research: what 162 searches returned from Shopify's agent catalogWhat 162 shopper searches returned from Shopify's agent catalog on October 26, 2026: 1,620 listings, which facts were present or blank, worked examples and a mistake we made. https://cartforagents.com/research/ai-product-data-unit-normalizationProduct data unit normalization: how we mistook -30 °C for a missing ratingReconstruct our minus-30 normalization failure with raw passages, expected matches and regression cases before labeling a product fact absent. https://cartforagents.com/research/ai-shopping-seller-concentrationSeller concentration in AI shopping results: count products, variants and sellersTen results can be one seller: a recount of dog-boot, backpack and tent shelves in Shopify's agent catalog by product ID, variant and seller. https://cartforagents.com/research/ai-visibility-score-methodologyAI visibility score methodology: why we retired our own scoreRebuild the failure of our first gap score using missingness, normalization errors and component contribution checks rather than a replacement grade. https://cartforagents.com/research/article-titles-agent-reading-simulationWhich article titles AI agents cite: a simulation across Claude, GPT and GeminiPlain titles were cited almost every time, clever ones mostly weren't: a simulation of which article titles Claude, GPT and Gemini models cite and choose to read. https://cartforagents.com/research/missing-product-review-dataMissing product review data: a blank rating is not a zeroAudit absent, null, explicit-zero and positive review states with field paths and denominator rules before calculating summaries. https://cartforagents.com/research/shopify-catalog-vs-chatgpt-resultsShopify Catalog vs ChatGPT results: how to report a catalog findingUse a proxy-to-outcome checklist to distinguish returned Catalog facts from live assistant retrieval, recommendations, seller choice and orders. https://cartforagents.com/research/ucp-catalog-native-checkout-researchNative checkout in Shopify Catalog: all 1,620 results said no, and every one had a buy linkRecount first-variant checkout flags in the tested unpersonalized context and separate a false flag from all live purchase routes. https://cartforagents.com/research/vague-specific-product-facts-simulationVague vs specific product facts: simulated picks by Claude, GPT and Gemini modelsVague listings were almost never picked; factual ones won when they answered the shopper's brief. Pick counts across Claude, GPT and Gemini models, by brief and product. https://cartforagents.com/research/quiet-humidifiersQuiet vs 21 dB: which humidifier facts reach AI shopping agentsEvery noise statement on a humidifier shelf in Shopify's agent catalog, traced to the field it came from: which listings gave a decibel figure and which only said quiet.