The Challenge
FoodDash ran food delivery in 12 Brazilian cities and had 4 million people opening the app every month. Orders were steady, but the team saw the basket wasn't growing — users ordered lunch and left. They wanted to add bakeries, grocery, and flowers, the categories that actually lift order frequency. The problem was brutal: every one of those categories was a different backend, a different catalog schema, a different ops relationship. "We priced out building grocery and it was a two-year, multi-team project," said Paulo Andrade, FoodDash's VP of Marketplace. "To sell someone a loaf of bread, we'd have to become a supermarket first."
The Solution
Cross Mini App's one-to-many distribution let FoodDash skip the build. The team dropped in the universal SDK once and opened the catalog to mini-programs already running on the Cross Mini App network — local bakeries, corner groceries, and flower shops that had built their own storefronts. Within a quarter, 200+ merchant mini-programs were live inside FoodDash, each owned and updated by the merchant. FoodDash never touched a single inventory feed.
The Results
| Metric | Before | After | Change |
|---|---|---|---|
| Non-food GMV / month | $1.1M | $2.0M | +85% |
| Merchant onboarding time | 6 weeks | 3 days | -90% |
| Monthly active merchants | 1,800 | 4,300 | +139% |
| Orders per user / month | 3.4 | 5.5 | 1.6x |
The non-food categories did what the team hoped: they raised order frequency. A user who came for a Thursday lunch started adding Friday groceries and Sunday flowers, and the app went from a once-a-week habit to a daily one. Merchants, meanwhile, got an audience they could never buy — FoodDash's 4M users — without building their own delivery app.
Key Takeaways
- Adjacent categories don't need adjacent engineering: FoodDash reached 200+ merchants by distributing existing mini-programs, not by staffing a grocery org.
- Onboarding collapsed from weeks to days: Because merchants bring their own catalog, FoodDash reviewed a storefront, not a data pipeline.
- Frequency is the real prize: Non-food GMV grew because flowers and groceries turned a weekly habit into a daily one.
- Merchants get the audience they lack: Local shops tapped 4M users without ever building their own app.
"We stopped trying to be a supermarket," Andrade said. "We just gave the supermarkets a door."