COS, the premium sibling of H&M, opened 20 new North American stores in the past 18 months and positioned itself as a direct competitor to J.Crew, Aritzia, and Banana Republic, according to WWD. The brand did not rely on a single channel. It deployed owned retail, e-commerce, and strategic wholesale partnerships simultaneously, executing a coordinated omnichannel push that smaller brands typically avoid because it looks expensive and complicated.

The mechanics were deliberate. COS opened flagship stores in high-traffic urban markets — New York, Los Angeles, and Toronto among them — to establish brand credibility and serve as physical proof points. At the same time, the brand expanded its direct-to-consumer e-commerce operation across the region, ensuring that consumers who discovered the brand online or in-store could buy through whichever channel they preferred. The third leg was wholesale partnerships with select department stores and specialty retailers, giving COS distribution in markets where it had no physical presence yet. All three channels reinforced each other: the stores drove brand awareness, e-commerce captured impulse traffic, and wholesale partnerships filled geographic gaps.

This worked because the brand solved the coordination problem that kills most omnichannel strategies. Retailers typically launch one channel at a time — open a store, then build e-commerce, then maybe try wholesale. By the time the third channel goes live, the first two are already stale or the brand narrative has shifted. COS avoided that by launching all three in the same 18-month window, with consistent product assortment, pricing, and visual identity across every touchpoint. A customer who saw a coat in a Toronto store could order it online in the same session. A buyer who discovered the brand through a department store partnership could later shop the full collection on COS's own site. The brand captured the sale regardless of entry point, and each channel became a discovery mechanism for the others.

A small physical-product brand can steal this play on a modest budget by running a compressed omnichannel test over 90 days. First, identify one geographic market where you already have some traction — email subscribers, past customers, or strong social engagement. Second, secure a small retail presence in that market: a pop-up, a consignment deal with a local shop, or a booth at a monthly market. The goal is a physical location customers can visit, even if it is temporary. Third, launch or refresh your e-commerce site with geo-targeted ads and local influencer partnerships in that same market, driving traffic from the retail presence to the online store and vice versa. Fourth, approach two or three local retailers or boutiques and offer a wholesale trial: consignment terms, net-30 payment, or a small initial buy with a reorder incentive. Stock them with the same hero SKUs you are featuring in your pop-up and online. Run all three for 90 days, then measure which channel drove the most first-time customers and which drove the highest lifetime value. The total cost: booth or pop-up rent, modest paid ads, and product on consignment. No need for 20 stores. You are testing the coordination mechanism, not the scale.

The broader pattern is that omnichannel works when the channels launch together and point at each other. If your retail presence sends people to your site, your site sends people to your wholesale partners, and your wholesale partners send people back to your owned retail, you have built a self-reinforcing loop. If each channel operates in isolation, you have built three separate businesses that compete for the same customer.