Online Sales Are Headed for $1.8 Trillion. Stores Are Not Going Anywhere Either.

A new Forrester report projects that U.S. e-commerce sales will reach $1.8 trillion by 2030, representing 29 percent of total retail sales. The remaining 71 percent, roughly $4.4 trillion, will still move through physical stores. Total U.S. retail sales, excluding automotive and gas, are forecast to grow from $5.2 trillion in 2025 to $6.2 trillion over the same period.

Both numbers hold up under scrutiny. Online sales are growing steadily, driven by a younger consumer base entering the workforce, improvements in fulfillment, and the early emergence of AI-assisted shopping. Physical retail, meanwhile, has shown more resilience than its critics expected, with store traffic proving far stickier than peak-pandemic forecasts suggested it would be.

What the headline figures do not capture is how the structure of the purchase journey is changing beneath the channel split, and what that means for the brands competing inside both environments.

The Share Story Obscures the Structure Story

Physical retail’s numerical dominance is real and worth understanding clearly. But the reasons consumers keep returning to stores have shifted in ways that matter for how retailers and CPG manufacturers invest.

McKinsey’s ConsumerWise research, drawn from a 2025 survey of more than 25,000 consumers across 18 countries, found that consumers are spending more of their discretionary time online and alone, a behavioral shift that, counterintuitively, has not collapsed in-store sales. What it has done is change why people go to stores. The trip is increasingly intentional, the basket more deliberate, and the comparison shopping that used to happen at the shelf now largely happens beforehand, on a phone or a laptop. The practical implication of that finding is that the store is becoming less a place of discovery and more a place of completion, where a purchase that was already largely decided gets finalized.

For CPG brands, this restructuring of the shopping journey carries a specific implication. If the decision is made before the consumer walks through the door, then the contest for that decision is happening in digital environments the brand may not control, may not fully understand, and may not be optimizing for. Physical shelf placement still matters enormously, but its power is increasingly derivative of digital discovery that preceded it.

Gen Z and the Omnichannel Misread

One of the more durable myths in retail strategy is that Gen Z is primarily an online generation. The data is more complicated. McKinsey’s ConsumerWise research characterizes Gen Z as financially cautious, paradoxically willing to splurge on specific categories, and deeply pragmatic about where and how they shop. McKinsey’s broader analysis of Gen Z grocery behavior, published in early 2024, found that groceries ranked as the top category where Gen Z and millennials planned to increase spending, ahead of travel, dining, and fashion. These are not the purchasing patterns of a generation that has migrated entirely to digital channels. They are the patterns of a generation that uses digital tools to prepare for physical purchases.

That preparation shapes how Gen Z moves through the retail environment. They discover products on social platforms, validate through peer reviews, and walk into stores with a high degree of prior conviction about what they intend to buy. For retailers and CPG manufacturers, this creates a specific planning problem. If Gen Z discovers a product online and then walks into a store expecting to find it, the failure mode is not the store. The failure mode is any break in the chain between digital awareness and physical availability.

Deloitte’s 2026 Retail Industry Outlook, based on a survey of 330 senior retail executives conducted between October and November 2025, found that uncertainty around technology investment and macroeconomic headwinds, including tariff pressures on imported goods, are already complicating the capital allocation decisions needed to keep those chains intact.

Deloitte’s ongoing ConsumerSignals research, which tracked approximately 9,000 U.S. consumers from September 2024 through May 2025, adds another layer: four in ten Americans currently exhibit three or more cost-conscious or deal-driven behaviors per month, a pattern Deloitte defines as value-seeking. This behavior is not confined to lower-income households. It cuts across demographics and reflects a consumer who approaches each purchase with more deliberation than two or three years ago. For CPG brands built on premium positioning, value-seeking behavior is not a cyclical condition to wait out. Deloitte’s analysis, which drew on approximately five million credit card transactions tracked over three years, found that brands perceived as delivering strong value relative to price gained measurable share, while those perceived as weak on that dimension lost it.

The Agent Problem Nobody Is Pricing In

Alongside the Forrester channel projections, a separate body of research points to a development that has received less attention in retail planning conversations: the rise of AI agents that shop autonomously on a consumer’s behalf.

McKinsey’s October 2025 analysis of the agentic commerce landscape estimates that AI agents could mediate between $3 trillion and $5 trillion in global consumer commerce by 2030. These are not purchases made after a consumer browses a website or walks a store. They are purchases executed by autonomous software systems that a consumer has authorized to search, compare, and transact on their behalf. OpenAI’s Operator, launched in January 2025, can navigate retail websites, assemble carts, and bring the user in only to confirm payment. Google has piloted an AI shopping mode that monitors prices and completes purchases automatically when a target price is reached. Payment networks including Visa and Mastercard are building APIs that allow verified AI agents to transact within pre-approved budget parameters.

McKinsey describes this shift as “a rethinking of shopping itself,” a move from discrete steps to a continuous, intent-driven flow managed by systems that do not browse pages the way people do. They parse structured data. They evaluate product metadata. They operate on signals that most brands are not currently engineering their content or product information architecture to send.

Gartner’s research offers a necessary counterweight. In a January 2025 poll of 3,412 executives, Gartner found that only 19 percent of organizations had made significant investments in agentic AI, while 42 percent described their investments as conservative. Gartner has also projected that more than 40 percent of agentic AI projects will be canceled by the end of 2027, citing escalating costs, unclear business value, and inadequate risk controls. Anushree Verma, a Gartner senior director analyst, described many current agentic AI projects as “early-stage experiments or proof of concepts that are mostly driven by hype.” Retailers and CPG companies would be right to avoid building strategy around a technology whose enterprise implementation remains largely unproven at scale.

The consumer-facing adoption curve, though, is a separate question from enterprise implementation. Consumers are already delegating shopping intent in measurable ways. NIQ’s 2025 Consumer Outlook research, drawn from a global survey of more than 17,000 consumers across 23 countries, found that 40 percent of respondents would accept a product recommendation from their AI assistant, and the same share said they would use AI to automate and speed up everyday shopping decisions. These figures represent stated intent, not observed transaction data, and should be read accordingly. But they signal a direction of travel that retailers and brand manufacturers cannot reasonably set aside.

When the Storefront Is an API

The strategic implication of agentic commerce for brand manufacturers does not map neatly onto existing marketing frameworks. If an AI agent is making or mediating purchasing decisions, it is not responding to creative advertising. It is not building an emotional relationship with a brand’s visual identity. It is not susceptible to in-store placement, promotional endcaps, or price-point psychology applied at shelf. It is evaluating structured product data, ingredient lists, nutritional information, price, reviews, availability, against a set of parameters the consumer has pre-established.

In this environment, the quality of a brand’s product data infrastructure becomes a competitive variable in the same way that shelf placement is a competitive variable in physical retail. Gartner predicted in January 2026 that 60 percent of brands will use agentic AI to deliver one-to-one consumer interactions by 2028. The brands positioned to benefit are those whose product information is clean, current, standardized across platforms, and structured in ways that AI systems can parse and compare reliably. Those that are not are operating on product content built for human browsers, rich in narrative and thin on machine-readable data, and that gap will be costly to close under time pressure.

CPG companies face a version of this challenge that their retail partners share only partially. A retailer owns the platform relationship with the consumer, however that platform evolves. A brand manufacturer does not. If the AI agent’s default is to select whichever product in a category has the strongest data signal, the cleanest structured feed, and the best-verified reviews, the brand that loses that evaluation has no endcap to compensate and no circular to fall back on.

The Lag Problem

The convergence of these pressures, value-seeking consumers, a more deliberate Gen Z shopper, and an emerging agentic commerce infrastructure, is happening on a timeline that retail planning cycles were not designed to match. Annual budget cycles, seasonal buying plans, and three-to-five-year capital roadmaps were built for a competitive environment where the variables changed slowly enough to observe and respond to before the next planning round.

NIQ’s 2025 Consumer Outlook, tracking CPG volume and value growth across 54 global markets, found that CPG inflation cooled significantly from its 2022-2023 peak but began creeping upward again between May 2024 and May 2025, by 0.3 percent. NIQ calculates that consumers today are spending the equivalent of $106 for what cost them $100 in 2023. Lingering price sensitivity, combined with tariff uncertainty heading into 2026, means the value-seeking behavior Deloitte documented is not a posture that will dissolve when the next set of quarterly results lands.

Deloitte’s 2026 Retail Industry Outlook found that retailers are already deferring supply chain investments in response to trade policy uncertainty. Those deferrals protect near-term margins. They also postpone technology integration, including inventory visibility, product data standardization, and API-ready catalog architecture, that will determine which brands remain discoverable as AI-mediated purchasing grows.

The Forrester forecast is a useful orientation. Retail is growing, online is growing faster within it, and stores are holding their ground. The more demanding question for senior leaders is whether their organizations are built to compete in a purchasing environment where the journey to a transaction is increasingly decided before anyone, or any agent, reaches the shelf.

Conversations On Retail

Conversations On Retail is a gathering place and resource center for retail and CPG executives, built to make it easier to stay current, discover the technologies and solutions shaping the industry, and connect with the people driving it forward.

We publish news, views, and reviews from staff editors, contributing experts, and trusted partners. Some articles are developed internally, while others are submitted by industry contributors or adapted from interviews and recorded conversations with industry leaders.