COS, the two-decade-old contemporary design line owned by H&M Group, is expanding across North America through a deliberate three-channel model: owned retail stores, direct e-commerce, and partnerships with independent retailers and department stores, according to WWD. The brand positions itself against J.Crew, Aritzia, and Banana Republic by combining H&M's supply muscle with a slower, design-led merchandise cadence and higher price points.

The mechanics are straightforward. COS operates its own flagship stores in major metros, runs a direct-to-consumer e-commerce site, and simultaneously places product in select third-party retail partners. This hybrid model lets the brand control brand experience in flagship markets while leveraging existing foot traffic and credibility in secondary cities where a standalone store would bleed cash. The partner retailers carry curated COS assortments, not the full line, maintaining scarcity while expanding geographic reach without lease risk.

The strategy works because it solves the unit economics problem that crushed fast-fashion's premium experiments. Opening owned stores in expensive North American shopping districts requires $500K-$1.5M per location in build-out and inventory before the first sale. COS sidesteps that in tier-two markets by riding on partner retailers' existing leases, staff, and local customer trust. The brand sacrifices some margin but gains distribution density without the fixed cost anchor. Meanwhile, the owned flagship stores in New York, Los Angeles, and similar anchors establish brand credibility and create content and styling authority that flows through to partner doors and online. The e-commerce channel captures customers between physical touchpoints and allows COS to test new markets with paid search and social before committing to brick-and-mortar.

For a small physical-product brand, the steal is running a tiered distribution model based on market maturity, not ego. Start with direct e-commerce to prove demand and build a customer file. Identify your top three to five metro markets by customer density from that file. In those markets, open owned retail — a showroom, pop-up, or micro-format store that functions as brand anchor and content studio. In secondary markets where you see 30-50 orders per month but not enough to justify a lease, approach independent retailers or regional chains that already serve your customer and propose consignment or net-60 terms with a curated SKU set of 12-20 hero products. You control imagery, POS materials, and staff training; they provide the lease, labor, and local credibility. Price the consignment margin at 40-45% to the retailer, which preserves your wholesale economics while keeping them invested. Use owned stores to generate styling content, user-generated content, and brand photography that arms your retail partners with ready-made social and email assets. Track which partner doors convert and expand selectively, treating retail distribution as performance marketing with a longer payback window.

The broader pattern here is that omnichannel is not about being everywhere. It is about sequencing distribution to match customer concentration and using each channel to de-risk the next. COS owns the expensive flagship real estate to build brand equity, then franchises that equity through partners in markets where the unit economics do not close on owned retail. The direct site captures everyone in between and feeds data back into the site selection and partner selection process. That cycle compounds. A small brand can run the same loop at lower altitude: prove online, anchor with owned retail in one or two cities, expand through selective wholesale where the customer file justifies it but the rent does not.