Glovo Partnerships Are a Poor Benchmark

The key point is that Glovo was proving a different grocery model, not the standalone dark store model that later quick commerce players were trying to make work. In the partnerships setup, Glovo used existing supermarket inventory, existing store footprints, and its own courier network to turn grocery into another category inside a broader marketplace app. That is lighter on capital and faster to launch, but it gives less control over assortment, in stock levels, and basket building, so it is a weak read through for what a vertically operated dark store can earn.

• The partnerships model is basically store pickup with a delivery layer on top. A customer orders from a chain grocer in the app, the order is picked from that retailer’s shelves, and couriers run the last mile. That avoids owning inventory, but the retailer still controls much of the margin structure and customer experience.
  1 sacra
• Dark stores flip that trade. Operators carry their own inventory in small neighborhood warehouses, which gives tighter control over stock, speed, and merchandising, but adds picking labor, spoilage risk, and local fixed costs. In mature form, that model can reach positive contribution margin, but only with enough daily order density and higher baskets.
  2 sacra 3 sacra
• That is why Glovo is not a clean benchmark. Its grocery economics were shaped by marketplace traffic and partner stores, while newer players like Getir, Gorillas, and Zapp were trying to prove whether owning the mini warehouse itself could create a better convenience business, even though most of the sector was still burning heavily at the time.
  1 sacra 2 sacra 3 sacra

Going forward, the market tends to reward models that own just enough of the stack to improve margin without taking on full supermarket complexity. The likely end state is fewer operators, denser networks, and a product mix that looks more like a high velocity convenience store than a full weekly grocery shop.