Unit growth hides a demand problem
Every chain here can grow its store count; the question the cluster exposes is whether the existing stores are still gaining customers. In 2025 the answer turned negative almost everywhere. Chipotle posted its first annual same-store decline since 2016 with traffic down all four quarters; Sweetgreen comps fell −7.9% (then −12.8% in Q1 2026); Chagee is eight quarters into a single-store sales slide. Revenue kept rising at most of them on new units — Chipotle still grew +5.4% — but the market looked straight through the expansion to the comp, and de-rated on it. The lesson the cross-section makes concrete: in restaurants, opening stores is capital allocation; same-store traffic is the business, and only one of them moves the stock.
The valuation spread is a bet on which comp recovers
Line the chains up by multiple and the disagreement is stark. CAVA trades at ~156× forward earnings, priced as the next Chipotle on a category it largely invented; Chipotle itself sits at ~29× after a ~46% slide; Sweetgreen at ~1.6× sales is priced near a solvency question, burning cash against a ~$130M cushion. The spread isn't really about today's profit — CAVA and Chipotle both run ~25% restaurant-level margins. It's a bet on whose negative comp is cyclical and whose is structural. CAVA's snapped back to +9.7% in Q1 2026; Sweetgreen's got worse. The market is paying 100× the multiple for the chain it believes re-accelerates.
China's chains prove the same point in reverse
Luckin and Chagee are the mirror image: explosive unit growth (Luckin added 8,708 net stores in a year to pass 31,000; Chagee runs 7,453) on top of collapsing per-store economics. Luckin's self-operated store margin swung from ~25% to 7% to 15% as the Cotti price war crushed the thin per-cup spread; Chagee's single-store sales fell ~40% over eight quarters even as total GMV grew. They are the cautionary extreme of the cluster's rule — you can out-store every rival and still be destroying unit economics, because in a saturating market new stores cannibalize old ones and the price war takes the rest. Growth in store count is not growth in the business.
Starbucks is what a turnaround costs
Starbucks, the largest by revenue at ~$37B across ~41,000 stores, is the incumbent mid-repair: operating margin fell 710bps to 7.9% as the new management's "Back to Starbucks" plan trades near-term profit for traffic (US comps did return to +7.1% in Q2 FY2026, but US transactions are still down). It frames the cluster's central tension cleanly — price increases papered over falling traffic for years, and unwinding that costs margin now. Across all six, the chains that pushed price to protect comps (Starbucks, Chagee) are the ones whose traffic broke first; the durable comps came from value and throughput, not price.
Where they agree — and where they split
All six accept that the 2025 slowdown was real and category-wide, and that same-store traffic, not unit count, is the scoreboard. They split on whether their own negative comp is a cyclical air-pocket or a structural verdict — CAVA and Chipotle lean cyclical (and Q1 2026 supports them), Sweetgreen and Chagee look structural, and the China chains are fighting a price war that makes the question moot until it ends. The demand for fast, affordable food isn't in doubt. Whether each specific concept can refill its existing stores — at a price that still earns a restaurant-level margin — is the only thing the market is pricing.