The US retail industry enters 2026 with cooler consumer demand, sticky service-sector inflation, and a channel mix that keeps shifting toward marketplaces and value formats. This guide pulls together the macro signals, channel data, technology shifts, and policy changes that retail and e-commerce teams actually need to plan the year, with a working playbook at the end.
In short
- Demand is normalizing, not collapsing: total retail sales growth lands in the low single digits in real terms, with discretionary categories the weakest.
- Marketplaces keep taking share: Amazon, Walmart, TikTok Shop, Temu, and Shein together account for the majority of incremental online demand.
- GenAI moves from pilots to P&L: merchandising copy, customer support, and demand forecasting are the first lines where retailers see measurable margin lift.
- Tariffs and nearshoring reshape sourcing: cost of goods on China-origin SKUs rises 8–15%, pushing Mexico, Vietnam, and India sourcing higher up the buy plan.
- Value and convenience win the year: discount grocers, club channels, and same-day delivery keep growing faster than the category average.
If you only have time for one read on the retail news landscape and how it shapes the global e-commerce industry, this is the place to start; everything below ties back to that bigger picture.
Why a 2026 outlook actually matters
Most retail forecasts read like horoscope columns: vague, optimistic, and impossible to act on. The point of a real outlook is the opposite. It should answer three questions for a category manager, a marketing lead, or a founder: what is the demand environment, where is share moving, and what investments earn their keep this year.
The retail industry outlook 2026 lands at a moment of unusual cross-currents. The labor market is normalizing but wages are still growing above pre-2020 trend. Inflation has cooled on goods but stayed warm on services. Interest rates have come down but mortgage rates are still high enough to keep big-ticket home goods soft. Marketplaces keep compounding share. Generative AI is moving from cost center to revenue lever. None of these are speculative; they all show up in current US Census retail trade data, public-company filings, and earnings transcripts.
The output should be opinionated. A planner who reads this and does not change a single line in the budget got the wrong document.
Key terms and definitions
The vocabulary around retail outlooks is full of overlapping shorthand. A quick reset makes the rest of the piece easier to read.
- Total retail sales: every dollar spent in retail trade and food services in the US, as reported monthly by the Census Bureau in the MRTS release.
- Core retail: total retail excluding autos, gasoline, and food services. This is the number most equity analysts watch because it strips out the volatile pieces.
- Comparable sales (comps): year-over-year sales growth at stores open at least 13 months. The single most-cited metric in retail earnings calls.
- GMV (gross merchandise value): total dollar value of orders processed through a marketplace, before refunds and before the platform takes its cut. Useful for marketplace size, misleading for profit.
- Take rate: the share of GMV a marketplace keeps as revenue. Amazon’s first-party plus third-party blended take rate sits in the mid-teens; TikTok Shop and Temu are still subsidizing theirs to buy share.
- Channel mix: the split of sales across store, online (own site), marketplace, wholesale, and licensed. Boards now ask for this on every brand review.
- Contribution margin: the margin left after variable costs (COGS, fulfillment, payment processing, returns, marketing). The metric that tells you whether a channel is actually paying its way.
If a term is missing, assume it is jargon that the writer did not understand themselves.
The macro backdrop for 2026
Three macro variables drive almost everything in a retail forecast: real disposable income, the savings rate, and the cost of credit. The 2026 setup is not catastrophic, but it is not loose either.
Real disposable income growth is positive but modest, in the 1.5–2.5% range for the year. The personal savings rate sits below its pre-2020 average, meaning households have less cushion than they did. Credit-card balances are at record nominal highs and 90-day delinquency rates have crept up, particularly in lower-income deciles. Mortgage rates have eased from their 2023 peak but remain a tax on housing-related categories: furniture, appliances, home improvement, and seasonal decor.
The practical read for retail planners: expect low single-digit nominal growth in total retail and food services, slightly negative in real terms for discretionary general merchandise, and positive in essentials, value formats, and food away from home.
What this means by category
| Category | 2026 direction | Why |
|---|---|---|
| Grocery and food at home | Modest growth, pricing-led | Volumes flat, price mix accounts for most of the dollar growth. |
| Discount and dollar formats | Above-average growth | Trade-down from middle-income households continues. |
| Apparel and footwear | Flat to slightly down | Closet refill cycle complete; weather and tariffs add volatility. |
| Beauty | Mid single-digit growth | Fragrance and prestige skincare keep working; mass color slows. |
| Home, furniture, appliances | Down low single digits | Housing-turnover proxy; recovers only when mortgage rates fall further. |
| Electronics | Slightly positive | PC and smartphone replacement cycle plus AI-feature pull-forward. |
| Restaurants and food service | Mid single-digit growth, pricing-led | Traffic flat to negative; ticket up on menu price increases. |
| Sporting goods, hobby, books | Flat | Post-pandemic boom fully digested. |
None of these numbers are guarantees. They are the central case. Build a low and high scenario around each with at least 200 basis points of range, then assign probabilities. A planner who only has a base case has no plan.
Channel shifts: where the next dollar comes from
The single biggest fact in US retail right now is that almost every incremental online dollar flows through a marketplace. Amazon still anchors the system. Walmart Marketplace has crossed a threshold where serious sellers cannot ignore it. TikTok Shop, despite regulatory questions, posts triple-digit GMV growth. Temu and Shein continue to mine the under-$25 price tier that legacy retailers abandoned. Etsy and eBay defend their niches.
For brands, this changes the math in three ways. First, customer acquisition cost on owned sites keeps rising as paid social and search costs reset higher. Second, the marketplace take rate (fulfillment, advertising, referral fees, returns) routinely exceeds 30% of GMV for a typical third-party seller, which is the new gross-to-net reality. Third, first-party retail media on Amazon and Walmart now functions as the most efficient performance channel available, which is why ad budgets keep migrating there.
Direct-to-consumer brands are not dead, but the model has evolved. The 2018–2021 playbook of paid-social acquisition into a Shopify funnel does not pencil. The brands that work now have a recognizable point of view, real wholesale distribution, and a marketplace presence that lets them harvest demand they cannot afford to create themselves.
Anyone trying to model channel mix in 2026 should also read the deeper take on how retail news shapes the global e-commerce industry today, which puts the marketplace shift in a multi-year context rather than treating it as a 2026 phenomenon.
Technology: where GenAI actually earns its keep
The 2024 and 2025 narrative on AI in retail was mostly demos. The 2026 narrative is P&L. Retailers are running structured programs in five places where the ROI is measurable within a quarter:
- Catalog and PDP copy at scale. Generated product titles, bullets, and descriptions, then human-edited at the top of the long tail. Conversion lift of 1–3% on the treated SKUs is now the published number from multiple public retailers.
- Customer service deflection. Tier-1 chat and email handled by tuned models, with human handoff on returns disputes and high-value orders. Cost-per-contact drops 30–60% where deployed well, with CSAT flat or slightly up.
- Demand forecasting and replenishment. Foundation models layered on top of existing demand planning tools, particularly useful for new-item forecasting where history is thin.
- Marketing creative production. Variant generation for paid social, lifecycle email, and onsite banners. Throughput goes up 5–10x; the cost saving is real but the bigger win is the testing velocity.
- Loss prevention and fraud. Computer vision in stores, anomaly detection on returns and account behavior. Shrink is a board-level issue at most large retailers, which makes this an easy budget approval.
Two things to flag. First, agentic commerce (AI agents that browse, compare, and buy on behalf of a consumer) is a real medium-term threat to the open web funnel. It is not material to 2026 revenue, but it is worth a strategy memo. Second, the build-versus-buy decision on most of the above is already settled in favor of buy. The marginal cost of integrating a managed model into an existing system has collapsed, and the talent to maintain custom stacks is expensive and scarce.
For tooling specifics, the companion piece on tools and vendors for industry in 2026 is the better place to dig in.
Supply chain, tariffs, and sourcing
Tariffs are the single biggest discretionary cost shock landing on US retail in 2026. The exact rate schedule will keep shifting, but the planning baseline is that landed cost on a typical China-origin general merchandise SKU rises 8–15% versus 2024. The pass-through to shelf prices is partial, which means gross margin compression for retailers that cannot move sourcing fast enough.
Three sourcing shifts are visible in the data already:
- Mexico for appliances, furniture, and auto parts, helped by USMCA and proximity to US distribution.
- Vietnam and Indonesia for apparel and footwear, with capacity constraints starting to bite.
- India for textiles, jewelry, and increasingly consumer electronics assembly.
None of these are turnkey. Vietnamese factory lead times have stretched. Indian quality systems vary widely by category. Mexican logistics infrastructure is improving but still concentrated on the I-35 corridor. The buyers winning in 2026 started qualifying alternatives in 2022 and 2023; the buyers starting now will pay a learning-curve premium for at least two seasons.
Ocean freight rates remain volatile but well off the 2021 peak. The bigger logistics story is the last mile, where same-day and next-day delivery expectations continue to compress economics for anyone who is not Amazon, Walmart, or Target. The category where this plays out most visibly is food, covered in detail in the analysis of grocery delivery economics and who actually makes money.
Payments, returns, and unit economics
Three line items have moved enough in 2026 to deserve their own paragraph in any board pack.
Payments. The Federal Reserve’s FedNow real-time payments network keeps expanding, and large merchants are starting to use pay-by-bank rails to bypass card interchange on high-ticket transactions. Buy now, pay later (BNPL) has settled into a mature share of checkout, particularly in apparel, electronics, and travel. Credit-card interchange remains the largest single payment cost; the long-running Visa and Mastercard settlement and its enforcement timeline matter to merchant margins more than most boards realize.
Returns. The cost of returns has crossed the threshold where retailers will openly charge for them. Restocking fees, return-shipping fees, and store-only return policies are now mainstream. The brands managing this best treat returns as a category-level merchandising problem (better size guides, fit technology, fewer SKUs with high return rates) rather than only a logistics problem.
Loyalty. Paid loyalty programs (Amazon Prime, Walmart+, Target Circle 360, Best Buy Plus) keep growing because they bundle delivery, returns, and exclusive content in a way free programs cannot. Free loyalty is fine for retention; paid loyalty is where the meaningful frequency lift is.
Policy and regulation: what to watch
Retail policy in 2026 is fragmented across federal agencies, state legislatures, and a handful of private antitrust suits. The items most likely to move the P&L:
- State privacy laws. More than a dozen states now have comprehensive privacy statutes in effect. The compliance burden is annoying; the meaningful business impact is on first-party data strategy, since cross-site tracking keeps eroding.
- Section 321 de minimis. Changes to the $800 de minimis threshold for direct-to-consumer imports would materially reshape the economics of Temu, Shein, and a long tail of cross-border sellers. The exact form of the change is still in flux; the direction (tighter, not looser) is clear.
- Card interchange and the Credit Card Competition Act. Any movement here flows straight to merchant gross margin.
- Robinson-Patman revival. The Federal Trade Commission has signaled renewed interest in price discrimination cases against suppliers favoring large retailers. Smaller retailers should watch this closely.
- Click-to-cancel and subscription disclosure. Subscription retailers should assume tighter disclosure and cancellation rules and design for them now rather than retrofit later.
For the underlying data behind any of these calls, the work of pulling together credible sources is laid out in the guide to retail industry data sources analysts actually trust. Government primary sources beat secondary research blogs every time, and the gap is widening.
Labor, store operations, and the wage-productivity squeeze
Wages in retail keep growing above the broader inflation rate. The minimum wage debate at the federal level has stalled, but a long list of states and major cities have moved their floors well above $15, and the largest retailers (Target, Costco, Amazon, Walmart) have set internal minimums that effectively act as the market clearing rate in most metros. The net effect is that hourly compensation is a persistent margin headwind.
The retailers managing this best are doing three things at once: investing in labor-scheduling software so that hours match traffic patterns more precisely, automating repetitive store tasks (shelf scanning, price changes, receiving) with robotics and electronic shelf labels, and redesigning roles so that experienced associates handle higher-value tasks (selling, fulfillment, customer issue resolution) while routine work shrinks. Self-checkout is being rebalanced after the shrink and customer-satisfaction issues of the 2020–2024 wave; the lesson is that fully unmanned checkout works only for small baskets in trusted formats, while staffed lanes return for larger baskets and tougher demographics.
Store fulfillment of online orders (ship-from-store, BOPIS, curbside) keeps growing as a share of e-commerce volume. The unit economics are good when basket size is high enough to cover the picking time and bad when it is not, which is why every meaningful retailer now has a minimum-order threshold and a delivery fee structure designed to push small orders into a more economic channel. Boards that benchmark store labor as a flat percent of sales miss the bigger picture; the right question is labor cost per fulfilled unit across all channels, weighted by basket profitability.
Common mistakes in building a retail outlook
Outlooks fail in predictable ways. The most common errors:
- Anchoring on last year’s growth rate. Retail is mean-reverting. A category that grew 12% last year because of a one-time tailwind will not repeat it.
- Confusing nominal and real growth. A 4% nominal sales gain with 3% category inflation is a 1% real gain, not a healthy number. Every line in the outlook should specify which it is.
- Treating marketplaces as a single block. Amazon, Walmart, TikTok Shop, Temu, and eBay behave nothing alike. Bundling them hides the actual share shifts that matter.
- Ignoring contribution margin. Revenue growth that costs more contribution dollars than it generates is value destruction. Channel-level CM is the right lens, not top-line GMV.
- Single-scenario thinking. The point of an outlook is to make better decisions under uncertainty. Without explicit low and high cases, the outlook is a forecast, not a plan.
- Confusing trends with cycles. Some moves (marketplaces taking share, AI productivity) are structural. Others (housing-related categories, BNPL penetration) are cyclical. Mixing them produces bad capital allocation.
- Hand-waving on policy. Tariffs, privacy, and interchange all have specific, modelable cost impacts. “We will monitor the regulatory environment” is not a plan.
Examples from US retail and e-commerce
Three concrete cases illustrate how the 2026 themes show up in real businesses.
A mid-sized apparel brand doing $80m in revenue, split 55% wholesale, 25% own e-commerce, 20% Amazon. The 2026 plan moves Amazon to 28%, opens a Walmart Marketplace presence at 4%, and holds wholesale flat in dollars (down in share). Marketing spend reallocates 30% from paid social to retail media on Amazon and Walmart. Sourcing moves 20 percentage points of the buy from China to Vietnam and India over the year, accepting a 90-day lead-time penalty in exchange for tariff insulation.
A regional grocery chain with 60 stores. The plan invests in electronic shelf labels (to manage faster price changes), expands the loyalty program to include a paid tier with free delivery, and consolidates two third-party delivery integrations into one to negotiate a better take rate. The board explicitly accepts a flat traffic forecast in exchange for higher ticket and basket margin.
A consumer electronics specialty retailer. The plan rebuilds the website search experience around a tuned generative model, leans into trade-in and refurbished as a margin-positive segment, and signs a service agreement to handle in-home setup for AI-enabled PCs and smart-home devices. Comp sales budget is plus 2%, but service and protection-plan attach drives most of the margin growth.
None of these companies is doing anything radical. They are reading the same data everyone else is and acting on it before the share shifts crystallize. The unifying thread is small, repeatable bets with measurable payback inside one or two quarters, rather than headline-grabbing transformations that take three years to validate and almost never survive a CFO change.
A fourth, less obvious example is the brand that decides not to chase incremental revenue this year. Some categories (premium home, certain discretionary apparel segments, second-tier marketplaces) genuinely do not pay back the investment required to compete. Choosing to harvest rather than invest in a soft year is a legitimate plan, provided the board agrees in advance and the team is not penalized for hitting a lower revenue target on purpose. Discipline beats ambition in a planning year like this one.
A working playbook for the year
If the team only has time to do five things in response to the retail industry outlook 2026, do these:
- Rebuild the channel-mix forecast from the bottom up, with explicit contribution margin per channel and an honest view of marketplace take rate.
- Model a tariff sensitivity at minus 5%, base, and plus 10% landed cost. Decide in advance which SKUs you raise prices on and which you delist.
- Pick two GenAI use cases with measurable payback inside 90 days. Stop the demos. PDP copy and tier-1 customer service are the safest starting points.
- Audit the loyalty program. Is the paid tier a real value exchange or a discount in disguise? If you have only a free tier, ask whether a paid tier is feasible.
- Tighten the data stack. First-party data is the only durable advantage left in marketing. Make sure consent capture, identity stitching, and downstream activation actually work end to end.
Everything else (re-platforming, store-format experiments, new-country expansion) is optional in a planning year like this one. Demand is too soft and cost is too volatile to fund discretionary bets that do not have a clear connection to the five items above.
For more context on how all of this fits into the broader retail news cycle and the structural shifts that will outlast any single calendar year, the pillar piece on how retail news shapes the global e-commerce industry remains the single best starting point on this site.
FAQ
What is the headline forecast for US retail sales growth in 2026?
Low single digits in nominal terms (roughly 2.5–4%), with most of the growth coming from price rather than volume. Real growth in discretionary general merchandise is likely flat to slightly negative; essentials, value formats, and food away from home continue to grow.
Are marketplaces really still taking share, or has that trend topped out?
Still taking share. The mix has shifted (Amazon’s growth has decelerated, Walmart Marketplace and TikTok Shop are accelerating), but the aggregate marketplace channel continues to grow faster than retail overall. Most brands underweight marketplaces in their channel plans by 2 to 5 percentage points.
How should we think about tariff exposure in 2026?
Build the plan around an 8–15% landed cost increase on China-origin general merchandise. Model price elasticity at the SKU level, identify the candidates for sourcing diversification, and accept that pass-through to consumers will be partial. Margin compression is the base case for retailers without a near-shoring strategy already in motion.
Is generative AI material to retail P&L in 2026?
Yes, in narrow places. PDP copy generation, tier-1 customer service deflection, and creative production all have measurable payback inside a quarter when implemented seriously. Other use cases (agentic commerce, personalized pricing, autonomous merchandising) are interesting but not material to this year.
Where should marketing budgets shift?
Out of broad paid social, into retail media networks (Amazon, Walmart Connect, Target Roundel, Instacart Ads) and into lifecycle marketing on first-party audiences. Brand investment remains important for category leaders; the cut is in mid-funnel performance spend that no longer pencils at current CPMs.
What is the single biggest risk to the base case?
A consumer credit shock. Card delinquencies are creeping up, savings rates are below trend, and any meaningful uptick in unemployment would push discretionary spending sharply lower. The base case assumes labor markets stay resilient; if they do not, the outlook turns from soft to ugly.
What sources should we trust for the underlying data?
US Census Bureau Monthly Retail Trade Survey, the Federal Reserve’s consumer credit and household debt releases, the Bureau of Labor Statistics for wages and CPI, and public-company filings for channel and category color. Secondary analyst notes are useful for synthesis but should never be the primary source for a planning number.
How often should we refresh the outlook?
A full rebuild quarterly, with monthly updates to the channel-mix and tariff scenarios. Anything less frequent is decorative; anything more frequent burns time you could spend executing on the plan.
Methodology note: figures cited are central-case estimates synthesized from public sources including the US Census Bureau Monthly Retail Trade Survey and the National Retail Federation retail sales tracker. Treat them as planning anchors, not point forecasts.