Comparison
Delivery apps vs your own storefront: what the commission is actually buying.
Marketplaces like Chowdeck, Glovo and FoodCourt put you in front of people who were never going to find you on their own. That is a real service with a real price. The question is not whether the price is fair — it is what you are still paying for once you already have the customers, the branches and the riders.
The short version
A delivery marketplace sells you demand, and if you need demand it is the cheapest you will ever buy. What goes with it is the thing that makes demand cheaper next time: the customer's name, the reason they ordered, and the price they think you charged. For a shop with no audience that is a fair trade. For a chain with branches people walk past every day and riders already on the payroll, it is renting back a neighbourhood you already own.
What a marketplace is genuinely good at
They bring customers you do not have
Being found is the hard part of selling online, and a marketplace solves it on your first day. A new store with no audience is not competing with the app — it has nothing yet to compete with. Anyone telling you otherwise is selling something.
A rider network you do not have to run
Hiring, equipping, insuring and scheduling riders is an operating business sitting inside your other one. If you do not deliver today, a marketplace is a fleet you can switch on this afternoon, and that is worth paying for.
Coverage past your own doorstep
An app reaches neighbourhoods where you have no branch and would never send your own rider. Your storefront only reaches as far as you can deliver, and no amount of software changes that.
What the commission costs you
- 01
A rate you do not set
Commission is the price of the demand, and the party selling the demand sets it. When it changes you are told, not consulted, and every order you have already built a margin around moves with it. A delivery fee in your own checkout is a number you chose and can change on a Tuesday.
- 02
You pay for the trip before the trip
A dispatch rider travels to the pickup point and then to the buyer. Both legs are in the economics, however they are labelled on the receipt — the customer is paying for a journey that happened before their order existed. When the goods start where the rider starts, that first leg is not a cost anyone has to carry.
- 03
The customer belongs to the platform
The name, the address, the order history and the reorder next Tuesday sit on the other side of the app. You supplied the goods, the shelf and the staff. When you want to tell that customer about a Saturday promotion, you pay again to reach someone who lives three streets from your door.
- 04
One number, and it has your name on it
Whatever the split between your price and the platform's, the customer sees a single total and attributes all of it to the shop on the bag. A delivery fee derived from a measurable trip — this distance, this weight, these surcharges — is a number you can explain. A blended one is a number you can only apologise for.
Why this bites harder in Lagos
Fuel moves, traffic is not the same at six as it is at ten, and a trip across the Third Mainland Bridge is not the trip the flat rate was built for. A marketplace absorbs that variance into one national-looking number because it has to — it is pricing for every merchant at once. A chain pricing its own deliveries can charge the Friday evening run differently from the Tuesday morning one, because it is only pricing for itself. The branches are already there. The riders, in most chains we meet, are already there too.
When you should stay on the marketplace
Leaving costs real money and real attention. Do not do it because a page on an agency website told you to. Three cases where the app is the right answer and we will say so:
Nobody is looking for you yet. If your own channels bring no traffic, the commission is buying the only demand you have, and building a storefront first is building a shop on an empty street.
You do not deliver and do not want to. Running a fleet means shifts, insurance, maintenance and somebody accountable at 7 p.m. when a rider does not show. That is a business decision, not a software one.
Your online volume is small. Below a certain number of orders, commission costs less than the staff time to run the alternative, and the honest answer is to leave it alone until the volume argues otherwise.
What running it yourself actually takes
A store worth sending people to
Catalogue, stock that reflects the shelf, payment that works on the first try. The app set an expectation and your storefront is measured against it from the first order.
Riders, and everything attached to them
People, shifts, bikes, fuel, insurance and a plan for the evening one of them does not arrive. Most chains we meet already have this. If you do not, this is the part to solve before the software.
Somewhere for the demand to come from
This is the one that is skipped. Signage in branch, SMS and WhatsApp to people who already shop with you, search, and a reason to order direct. The marketplace was doing this, and when you stop paying it, it stops.
Not sure which side of this you are on?
If you already have branches and riders, the maths is probably closer than it looks. If you do not, we will tell you to stay where you are. Either answer takes one conversation.