Online shopping changing general merchandise retail
This research explores how general merchandise retail is changing due to online shopping. It will examine shifts in shopping behavior, retail operations, and competitive dynamics driven by e-commerce.
Last update Sep 12, 2026, 1:01 PM EST
Intelligence Brief
The current state and what matters now
Actors
The field is still led by Amazon, Walmart, Target, Costco, and a long tail of marketplace sellers and private-label operators. The actor set is widening around the layers that shape discovery, routing, and checkout: AI shopping assistants, cross-merchant commerce layers, retail media networks, commerce data providers, membership ecosystems, community validation platforms, and social commerce platforms.
- Amazon remains a major shopping layer, but the signal is now broader than assistant-led research: it is also testing merchant routing, price-history automation, and delegated buying.
- Walmart remains central as a fulfillment and assortment operator, with marketplace breadth and cross-border expansion reinforcing its role as a scale platform.
- Target remains important for AI-native discovery and non-merchandise monetization, with visual search and personalized shopping flows gaining attention.
- Google is becoming more prominent as a cross-surface shopping orchestrator, with Universal Cart and commerce protocol work pushing it toward a shared transaction layer across Search, Gemini, YouTube, Gmail, and Google Pay.
- TikTok Shop continues to strengthen discovery-led commerce where creator content and live shopping shape intent before search.
- Reddit remains a validation layer where shoppers check recommendations before buying.
- Shopify remains strategically important as commerce infrastructure because merchant feeds and checkout rails can be exposed to external AI and shopping surfaces.
Moves
- Cross-surface shopping is intensifying fastest, with shopping increasingly spanning search, chat, video, email, and payments rather than staying inside one retailer site.
- Shared cart infrastructure is emerging as a structural layer, making cart persistence and checkout routing more portable across merchants and surfaces.
- AI-mediated discovery remains active, but the emphasis is shifting from pure recommendation to routing, handoff, and transaction completion.
- Delegated shopping is strengthening through price history, alerts, and auto-buy, though consumer oversight remains important.
- Marketplace growth remains a primary engine for general merchandise, with third-party assortment used to widen selection and defend price competitiveness.
- Discovery-led commerce is still emerging, with creator content and entertainment feeds acting as a first touchpoint rather than a supporting channel.
- Community validation remains a recurring checkpoint, especially when shoppers use forums to verify AI recommendations.
- Retailer-owned handoff flows are still emerging, where external surfaces route shoppers into retailer-controlled assistants or checkout paths.
- Trust tooling continues to expand inside shopping assistants, including scam verification, price history, and automated alerts.
Leverage
Advantage now comes from controlling the full commerce loop: discovery, trust, assortment, fulfillment, monetization, and increasingly checkout routing. The strongest players combine traffic, first-party data, inventory density, delivery reliability, and interface ownership.
The newest leverage point is cross-surface intent capture: whoever controls the shared cart, protocol, or assistant layer can shape demand before the shopper reaches a retailer’s site. Control over product feeds, identity linking, catalog quality, and inventory accuracy is now a source of bargaining power.
A second leverage point is marketplace control. Retailers that can expand third-party assortment without losing trust can widen selection, improve price competitiveness, and monetize the interface through media and services.
A third leverage point is interface control. Retailers and platforms that can separate discovery from checkout without losing the customer can monetize routing, not just transactions. Visibility inside AI search, social feeds, and shopping surfaces is becoming a new form of shelf space.
A fourth leverage point is price confidence. Price history, alerts, and auto-buy rules reward merchants that can sustain trust over time, not just win a single click.
Constraints
- Thin margins still limit how much price competition and free shipping can be absorbed.
- Fulfillment costs remain structurally high for bulky, low-value, or high-return general merchandise.
- Product-data quality is now a hard constraint: if catalogs, attributes, pricing, or availability data are wrong, AI search and marketplace conversion degrade quickly.
- AI adoption is outpacing execution, creating a gap between strategic intent and operational readiness.
- Consumer trust remains incomplete: shoppers appear more willing to use AI for help than to let AI decide and buy without oversight.
- Human validation is strengthening as a check on AI, which slows full delegation of purchase decisions.
- Merchant defenses against bots remain a friction point, because retailers must distinguish helpful agents from malicious automation.
- Marketplace abuse remains a risk, including counterfeit listings, scam goods, and unauthorized sellers.
- Checkout fragmentation persists even as cart layers improve, so discovery, comparison, and payment still often happen on different surfaces.
- Regulatory interface requirements are becoming more concrete, including visible cancellation and withdrawal functions for some cross-border retailers.
Success Metrics
Success is increasingly measured by profitable digital penetration, not just online sales growth. Key metrics include gross margin after fulfillment, repeat purchase rate, order frequency, basket size, conversion rate, and customer lifetime value.
- Cross-surface metrics: routed checkout completion, off-site add-to-cart rate, and the share of demand that begins on one platform and completes on another.
- AI-era metrics: assisted conversion rate, recommendation accuracy, search-to-purchase time, share of traffic influenced by agents, and share of orders completed through conversational or delegated surfaces.
- Discovery-led metrics: feed-to-cart rate, creator-driven conversion, watch-to-cart rate, and the share of sales originating in short-form video or livestreams.
- Price-guided metrics: share of purchases triggered by price alerts, target-price rules, or auto-buy behavior.
- Marketplace metrics: seller quality, take rate, assortment depth, and trust signals.
- Fulfillment metrics: on-time delivery, pickup adoption, return rates, and inventory turns.
- Community metrics: validation click-through, forum-assisted conversion, and the share of shoppers who seek second opinions before purchase.
Underlying Shift
The deeper shift is from a store-centric distribution model to a data- and AI-orchestrated commerce system. Online shopping has already changed general merchandise retail by making assortment, pricing, logistics, and media continuously adjustable. The new phase goes further: shopping is becoming mediated by assistants, recommendation engines, membership ecosystems, creator networks, community forums, and unified operating layers that blur the line between browsing, buying, and fulfillment.
The retailer is less a shelf owner and more a platform operator coordinating demand across digital interfaces, stores, and delivery networks. In this model, the store is a node, the app is a control surface, and AI is becoming the front door and, increasingly, part of the checkout layer.
The latest signals suggest the system is moving from AI-assisted shopping toward cross-surface commerce operations, with product data, agent compatibility, trust controls, and demand timing becoming part of the operating core. A newer layer is shared cart and transaction infrastructure, where shopping can begin in one surface and complete in another without losing state.
A second emerging layer is marketplace-led retail: third-party assortment is no longer peripheral, but a structural growth engine for general merchandise.
A third layer is discovery-led commerce, where social feeds, creators, short-form video, live shopping, and community validation increasingly act as the first shopping surface rather than a marketing side channel.
A fourth layer is regulated interface design, where cancellation, withdrawal, and post-purchase actions are becoming part of the commerce UX itself.
Current Phase
The market is in a late adoption, early transformation phase. Omnichannel is no longer novel; it is table stakes. The next competitive wave is about who can operationalize unified commerce, cross-surface shopping, marketplace-led assortment, and regulated interface flows without destroying margin or trust.
The profit pool is still contested among retailers, marketplaces, brands, creators, and commerce media businesses, but the battle is shifting toward who owns the customer interface, the AI layer, and the transaction rules. The winners will be those that turn complexity into a simpler customer experience while also reducing internal friction and improving inventory truth.
The evidence now suggests online shopping is not just supplementing general merchandise retail; it is increasingly setting the operating logic for it. Attention appears to be shifting from whether AI matters to where the shared cart and transaction layer sits, from whether marketplaces help to how they become the default growth engine, and from seasonal promotion to platform-controlled demand timing.
What to Watch
- Cross-surface cart and checkout, especially whether carts can persist across search, video, email, chat, and marketplace portals.
- Agentic and conversational commerce adoption, especially whether AI assistants become a meaningful source of traffic and conversion.
- Delegated shopping behavior, especially whether price alerts, auto-buy, and assistant-led replenishment become routine in general merchandise.
- Discovery-led commerce scale-up, especially whether creator-led and feed-led shopping become repeatable, high-volume channels in general merchandise.
- Retailer-owned handoff models, especially whether external LLMs increasingly route shoppers into retailer-controlled assistants.
- Marketplace operating models, especially whether marketplace becomes a core assortment and growth system rather than a side channel.
- Merchant acceptance of AI checkout, including whether retailers monetize assistant traffic or keep blocking it as bot risk.
- Assisted local shopping, including whether AI calling and verification become standard for high-consideration general merchandise.
- Marketplace enforcement, especially whether fraud and counterfeit controls become a standard operating layer.
- Price-alert and auto-buy usage, including whether shoppers trust delegated purchase execution for repeatable categories.
- Unified commerce execution, especially whether retailers can truly collapse channel silos for customers and operations.
- Commerce media integration, including whether on-site, in-store, and offsite media become one measurable system.
- Inventory and catalog accuracy, since AI ordering and availability tools only work if product data stays clean.
- Community validation, especially whether Reddit-like checkback behavior remains a durable step in AI-assisted purchase journeys.
What's new
Latest brief updates
What’s new: The brief was updated to reflect a stronger shift toward cross-surface transaction infrastructure, especially Google’s Universal Cart and commerce protocol, which now make shared cart, checkout, payments, and post-purchase flows a more central organizing layer. It also sharpened the role of AI shopping in general merchandise by adding recent signals from Amazon, Shopify, and Target, while keeping the earlier interpretation that marketplace growth, community validation, and discovery-led commerce remain important. No updates since the previous Brief were needed for the broader phase view beyond emphasizing that the center of gravity is moving from AI-assisted browsing to routed, cross-surface commerce operations.
Dominant Themes
High-density signal formations
Loading cluster map
Aggregating signals by recency and strength
Fastest-Rising Themes
Themes showing the strongest momentum
Loading cluster history
Reading snapshot progress over time
Analysis
Interpretation of what’s changing
Commerce Is Becoming a Resume Button
Full analysis summary: AI shopping is quietly changing the unit of competition. The old game was winning the click, then converting it before the shopper drifted away. The new game looks more like keeping a tab open in the customer’s mind — and in the system. Google’s Universal Cart, buy buttons inside AI Mode and Gemini, Target’s continue-browsing and Buy Again features, Amazon’s Alexa shopping assistant, and Walmart’s embedded experience in Gemini all point in the same direction: shopping is turning into a persistent workflow. Discovery, comparison, carting, and checkout are no longer separate rooms. They are becoming one long hallway, with the shopper able to step out and come back without starting over. That matters because it changes what “loyalty” means. If the assistant remembers the cart, the preferences, the prior search, and the unfinished intent, then the retailer or platform that owns that memory layer can reduce friction every time the shopper returns. In practice, that is a retention advantage disguised as convenience. The winner is not just the best storefront; it is the best resume button. There is a second-order effect here too: AI makes repeat intent easier to capture than fresh demand. Target surfacing past purchases and relevant deals is not just personalization; it is a mechanism for turning old intent into new orders with less effort. That should improve conversion efficiency, but it also means more leakage risk for merchants that cannot preserve context across devices, sessions, or channels. The uncertainty is that this only works if shoppers trust the assistant and if the underlying commerce plumbing is interoperable. A persistent workflow is powerful, but only when the cart survives the handoff. If context breaks, the funnel reappears.
Retail’s New Front Door Is Machine-Readable
Full analysis summary: Retail discovery is quietly moving upstream. The store shelf is no longer the first place a shopper “sees” a product; increasingly, it is the place where a decision already made elsewhere gets executed. The real gatekeeper is becoming the layer that AI can parse: structured metadata, clean product feeds, conversational assistants, and interoperable commerce protocols. That is why the signals matter together. NRF-linked commentary about clean data and smarter search, Amazon’s feed-based distribution, Google’s protocol spanning discovery through post-purchase, and LinkedIn-style guidance on structured metadata all point to the same mechanism: if a product cannot be read by machines, it is less likely to enter the shortlist in the first place. It is like moving from a brightly lit storefront to a customs checkpoint—what matters is not only what you sell, but whether the system can classify it fast enough to let it through. The strategic implication is uncomfortable for retailers that still think of AI as a front-end feature. The contest is not just for better onsite search; it is for inclusion in the pre-cart decision layer. That layer is where comparison happens, where “consideration” is formed, and where media efficiency is increasingly won or lost before a shopper ever lands on a product page. There is a catch, though. Structured data and AI visibility do not guarantee demand; they only improve the odds of being legible. If the underlying product, price, or proposition is weak, machine readability just makes the weakness easier to ignore at scale. And the system is still uneven: some categories and platforms are far more AI-ready than others, so the shift will be lumpy rather than universal. Still, the direction is clear. Retail is becoming a two-stage system: machines decide what gets considered, and the store decides what gets completed.
AI Shopping Is Turning Product Data Into a Planning Weapon
Full analysis summary: Retailers used to think of commerce data as a scoreboard: useful for proving what happened after the fact. That frame is getting too small. As shopping assistants, machine-readable catalogs, and commerce protocols spread, the same product data that helps an AI recommend something also becomes the input for what gets stocked, priced, promoted, and measured next. The shift is subtle but important: AI is not just changing discovery, it is tightening the loop between discovery and decision. If an assistant can compare products, check price history, and track items across the web, then the retailer site is no longer the endpoint of the journey. It becomes a verification layer, and the underlying data becomes the thing that decides whether the shopper trusts the result. In that world, structured catalog quality is not a back-office hygiene issue; it is visibility infrastructure. That creates a new power center. Retailers and brands with clean, interoperable commerce signals can feed those signals back upstream into assortment and media planning faster than competitors can. The advantage is not just better attribution. It is lower decision latency: fewer delays between what customers are doing and what the business changes in response. Retail media’s move from inventory and revenue toward effectiveness and customer value fits that pattern. Measurement is becoming the front edge of planning, not the back edge of reporting. But there is a catch. The market is also describing signal fragmentation as the main friction, which means many players will have more data but less usable coherence. More dashboards do not equal more control if the signals cannot be compared, trusted, or operationalized across partners. Google’s commerce protocol points toward a more connected stack, but the industry is not there yet; a lot of product data is still too messy for AI layers to interpret consistently. So the real contest is not over who gets the most retail media impressions. It is over who can turn commerce signals into a closed-loop operating system before everyone else is still stuck reading the receipt.