For shops and restaurants
Most businesses end up doing both, for different jobs. The platforms bring you customers who have never heard of you, and you pay a commission on every order for that. Your own delivery keeps the margin and the customer details on the orders that were coming to you anyway. If nobody knows your name yet, start with the platforms. If people in your city already know your shop, the regulars are where your own channel pays for itself, because those orders did not need buying.
| Listing on Swiggy or Zomato | Running your own delivery | |
|---|---|---|
| Reach | Puts you in front of people who have never heard of you. | Reaches the people who already know your shop. |
| Cost per order | A commission on every order, at the rate in your agreement. | Your rider and platform costs, which you control and which do not scale with order value. |
| Customer details | Stay with the platform. You do not get the phone number. | Stay with you. |
| Pricing and offers | Platform rules, and pressure to discount against everyone else on the app. | You set it. |
| Who the customer belongs to | The app. They opened it, not your shop. | You. |
| Best for | Being discovered, and one-off orders. | The regulars who order every week. |
This trips people up, so it is worth being clear. Zomato and Swiggy are marketplaces: they list your shop, send you orders and take a commission. Blinkit, Swiggy Instamart and Zepto are not marketplaces. They buy stock, hold it in their own dark stores and sell it themselves. A supermarket cannot list on Blinkit however much it wants to, because Blinkit is already selling the same rice and the same onions down the same street. FreshToHome works the same way in fish and meat. So if you run a grocery or a meat shop, quick commerce is not a channel decision at all. It is a competitor, and your own customers are the only thing it cannot take from you.
We are not going to quote you a percentage, because your rate is in your agreement and you know it better than we do. Take it, and apply it to the orders that came from customers who already knew your shop. Those are the orders you paid a commission on to reach people you had already reached. For an established supermarket or restaurant that share is usually the bigger one, and it is the part your own channel takes back.
In a Kerala city the same names come back every week. That is true for us in Thodupuzha and it will be true for your shop. A customer who orders from you four times a month through an app is four commissions on a relationship you already had. The same customer ordering direct is the same four orders at your own cost, with their number in your own records.
Discovery. The platforms are where people look when they do not know who to order from, and nothing you build replaces that. You also take on work you did not have before: riders, delivery zones, someone answering the phone when an order is late. That work is real, and it is most of what we help with.
Less than most people assume once the customers already exist. We run Thodupuzha with about five riders at a time, a 10 km radius, food out in 30 to 45 minutes and everything else within a few hours. The platform side is software and can be ready in weeks. The operations side is where the six years went, and it is the part you should not have to learn twice.
If you do not have a customer base yet, your own delivery channel has nobody to serve, and this page is not for you. Start on the platforms, build the regulars, then take them direct. Running your own delivery is a way to keep customers you already have, not a way to find new ones.
Questions
On the orders from customers who already knew you, usually yes, because you are not paying to reach someone you had already reached. On genuinely new customers the platforms are often worth their rate. That is why most businesses run both rather than choosing.
No. Blinkit, Swiggy Instamart and Zepto are not marketplaces. They hold their own stock in their own dark stores and sell it directly, so there is nothing to list on. For a grocery or meat business they are a competitor rather than a channel, and that is exactly why having your own customers matters.
Running both is common. What you are allowed to do is set by your agreement with each platform, so read that or ask them. We would not advise you on what is in a contract we have not seen.
There is no single number, and anyone who gives you one is guessing. It depends on your order values, how concentrated your customers are and how far you deliver. We look at that with you before you spend anything, and we will say so if the answer is that it does not add up yet.
Some will and many will not, which is why ordering has to work over WhatsApp and the web too. In a town where people already know your shop, a message is often the whole ordering flow. We build for that rather than assuming an app install.
Not necessarily, and how you structure it varies. What matters more is how many you run and how they are paid. We keep about five on the road at a time in Thodupuzha, deliberately, so the people working earn properly instead of splitting the same work more ways.
We have run a delivery service in Thodupuzha since 2020: more than 30,000 customers and over 5,00,000 orders from one mid-sized Kerala town, without expanding anywhere else. We also build and license the same platform to other businesses, so we have an interest here, and the honest version above is the one that has held up.
Also worth reading
Talk it through
Or see the numbers first in the Thodupuzha case study.