Online retail growth.
A company-level decision.
A real-source business brief. Original analysis, not a client engagement.
Executive judgment
Treat online retail growth as a reason to investigate a channel, not as proof that entering it will be profitable. Before committing to expansion, test whether your category, fulfillment economics and customer acquisition costs support the decision. A national trend cannot answer those company-level questions.
Evidence snapshot | historical release
The Census release dated August 18, 2026 reported Q2 U.S. retail e-commerce sales of $340.2 billion, seasonally adjusted, and a 17.1% share of total retail sales. It reported annual growth of 12.2% for e-commerce versus 6.7% for total retail. These estimates are nominal, not inflation-adjusted. [1]
Critical limitation
Census subsequently posted a September 28 notice that this release no longer contains the most up-to-date estimates; revised figures are scheduled for the November 19 release. This brief uses the August release as a dated evidence example, not a claim about the latest revised market size. FRED still displays the August-vintage 17.1% observation. [1, 2]
What this supports - and what it does not
The reported series supports examining online channels alongside other routes to customers. It does not establish demand for a particular product, customer willingness to pay, attainable market share, or the profitability of paid acquisition. It also does not mean physical retail is shrinking. Confidence is high in what the cited release reported; confidence in any company-specific opportunity remains unassessed.
Decision gates
1. Define one category and buyer. Specify the product range, geography and use case before estimating opportunity.
2. Compare the actual alternatives. Record delivered price, delivery promise, return terms and positioning for a bounded competitor set.
3. Build contribution economics. Include product cost, fulfillment, payment fees, returns and acquisition cost. Do not substitute gross revenue growth for contribution profit.
4. Run a bounded test. Agree the spend limit, learning objective and stop conditions before launch.
Evidence that would change the recommendation
Proceed to a larger test only if observed contribution after acquisition costs is positive at acceptable service levels and repeat demand is credible. Rework the offer if purchases depend on unsustainable discounts. Pause if shipping, returns or customer acquisition erase contribution. These are proposed decision rules, not measured results or universal thresholds.
A focused research assignment
A useful next brief would compare a defined set of competitors, document buyer and channel evidence, and identify the assumptions a pilot must test. Required client inputs would include the category, target market, current costs and decision deadline. No confidential client data or client engagement was used in this public demonstration.
Sources and method
[1] Census release and revision notice. [2] FRED ECOMPCTSA, the same underlying Census series. Reviewed October 31, 2026. No interviews, proprietary data or forecasts.
Discuss your business question ↗