Why Restaurant Owners Are Skipping Uber Eats and Building Their Own Delivery App

A common complaint from restaurant owners who use Uber Eats or DoorDash is this: the app brings orders, but 20–30% of every ticket goes to the platform before it even touches their bank account. That pain runs deep enough that "just build our own app" comes up constantly.
But here's the part that gets skipped in that conversation: leaving the aggregator doesn't automatically mean you keep that 20–30% for yourself. The aggregator was charging that commission because it was bringing you demand, including discovery, marketing, and an existing user base that already had the app installed. If you build your own platform, you still have to solve for all of that. You're not removing a cost; you're taking on a job the aggregator used to do for you.
So the real question isn't "can I avoid paying commission?" It's "can I generate enough of my own order volume that owning the platform is cheaper than what I'm currently paying in fees, once you count the cost of driving that demand yourself?" For a single-location restaurant with modest volume, that math often doesn't work. For a multi-location group, a cloud kitchen network, or someone trying to build a local aggregator for other restaurants in their area, it can work out very differently.
The bigger lesson is that the hard part isn't actually building the app. It's getting customers to use it. If you're testing the model, a pre-built option like Zipprr's Uber Eats Clone can let you test the platform side without committing to a large custom development project upfront. Then you can put your time and budget toward the part that actually determines whether the business works: acquiring and retaining customers.
Has anyone here tried this approach? I'd be interested in hearing what worked and what didn't.
https://zipprr.com/ubereats-clone/