For most restaurants with regulars, your own ordering website costs far less than Zomato and Swiggy commission. The apps keep 18–25% of every order, plus 18% GST on that commission. A basic ordering website in India costs roughly ₹40,000 to ₹1,50,000 to build and ₹2,000 to ₹3,000 a month to run, and Razorpay keeps about ₹9 on a ₹400 order. So a lean site pays for itself after about 370–530 direct pickup orders. That doesn't mean you should quit the apps. Keep Zomato and Swiggy for discovery and move your regulars to your own site.
I'm Mohit Goyal. I built a WooCommerce ordering system with delivery radius rules and automatic kitchen printing for Whitfield Service Station, a UK client. Below I run the same maths for an Indian restaurant. The India numbers come from published sources, not from one of my projects, so treat them as general guidance. If you're in the UK, I've written a separate post on takeaway websites vs Just Eat and Deliveroo.
What do Zomato and Swiggy actually take from each order?
Zomato and Swiggy take 16–30% of the order value as commission, most often 18–25%, and you pay 18% GST on that commission on top. Payment fees, extra levies, discounts you fund and in-app ads also come off the payout.
Business Standard reported in June 2025 that the two apps typically charge 16 to 30 per cent. POS vendors, who see a lot of restaurant contracts, give a tighter range. Petpooja puts Swiggy at 18–25%, and DineOpen quotes Zomato at 18–25% and Swiggy at 18–28%.
And the commission is only part of it:
- GST on commission. You pay 18% GST on the commission itself. Payment fees of about 2%, also taxed, are often added as well.
- Extra levies. Swiggy started charging some restaurants an extra ₹2–₹5 per Swiggy One order from 25 November 2025.
- Discounts and ads. Nobody publishes what restaurants spend on funded discounts and ads inside Zomato and Swiggy. Pull those figures from your own payout statements.
- Customer fees. Your customers pay too. Both apps raised their platform fee to ₹17.58 per order in 2026. A regular who orders from you directly skips that fee, which gives them an actual reason to switch.
Then there's the customer data. The apps own the relationship and mask contact details. That was one of the complaints in the NRAI case before the Competition Commission. If your best customer only ever orders through an app, you have no way to contact them yourself.
What does a restaurant ordering website cost in India?
A basic restaurant ordering website in India costs about ₹40,000 to ₹1,50,000 to build. A full agency site with ordering and reservations costs about ₹2,20,000 to ₹4,20,000. Running costs are around ₹2,000 to ₹3,000 a month, plus Razorpay's 2% + GST per payment.
The published ranges are wide because "ordering website" can mean a menu with a cart or a fully custom platform.
| Cost item | Typical range | Source |
|---|---|---|
| Build: basic ordering site | ₹40,000 – ₹1,50,000 | Crafton Labs |
| Build: full agency site with ordering and reservations | ₹2,20,000 – ₹4,20,000 | ColorWhistle |
| Hosting, domain and upkeep | ₹2,000 – ₹3,000 a month | Crafton Labs |
| Razorpay payment fee | 2% + GST per successful payment | Razorpay |
| Cloud kitchen printing (PrintNode) | Monthly plan, see PrintNode's current pricing | PrintNode pricing |
| On-demand delivery (if you don't have riders) | From ₹27 per drop, depending on distance | Borzo |
What pushes the build cost up? Mostly custom design instead of a good theme, a complicated menu with variants, add-ons and combos, and custom logic like delivery zones, time slots and printing. If you want to compare beyond restaurants, I've written a wider guide to WooCommerce website costs in India.
How many direct orders does it take to break even?
On a ₹400 average order, a ₹40,000 ordering website breaks even after about 370–530 direct pickup orders, and a ₹1,50,000 site after about 1,380–1,990. Covering ₹3,000 a month in running costs takes only about 28–40 direct pickup orders a month.
Here's what I assumed: a ₹400 average order, app commission at 18% and 25% with 18% GST, and Razorpay at 2% plus GST (₹9.44) on your own site. I left out the app's own payment fee, so these savings are on the cautious side. The delivery row subtracts Borzo's ₹27 starting price. Your real delivery cost depends on distance and will usually be higher.
| Scenario (₹400 order) | 18% commission | 25% commission |
|---|---|---|
| App keeps (commission + GST) | ₹84.96 | ₹118.00 |
| Own site keeps (Razorpay 2% + GST) | ₹9.44 | ₹9.44 |
| Saving per direct pickup order | ₹75.52 | ₹108.56 |
| Saving per direct delivered order (after ₹27) | ₹48.52 | ₹81.56 |
| Pickup orders to recover a ₹40,000 build | ~530 | ~368 |
| Pickup orders to recover a ₹1,50,000 build | ~1,986 | ~1,382 |
| Pickup orders to recover a ₹4,20,000 build | ~5,561 | ~3,869 |
| Pickup orders a month to cover ₹3,000 running costs | ~40 | ~28 |
The table makes a strong case for a lean build. If 10 regulars a day move to your site, a ₹40,000 build pays for itself in about two months once running costs are included. A ₹4 lakh build at the same volume takes more than a year. My advice: put the money into what gets people ordering again, like a fast checkout, saved addresses and printing that doesn't fail, and skip the showpiece design.
Should you leave Zomato and Swiggy completely?
Usually not. Zomato and Swiggy are where people who've never heard of you find you. For most Indian restaurants the right setup is to stay on the apps and move repeat orders to your own website.
I think the commission on a first order is a fair price for a new customer. Paying it on the twentieth order from the same person is where it hurts. Run both, and use each for what it's good at.
Check your price parity terms
Zomato reportedly dropped a clause that penalised restaurants for charging less for dine-in than on the app. That report covers dine-in prices, and Swiggy had not announced a similar change at the time of writing. Read your own agreement before pricing anything lower on your site, and compete on convenience if in doubt.
That comes from Business Standard's April 2026 report, which is about dine-in pricing. It doesn't say whether the change extends to prices on a restaurant's own ordering website. The NRAI has since asked the CCI for interim relief so these clauses don't come back. I'm a developer, not a lawyer, so take this as background, not legal advice.
How do you move repeat customers to your own site?
Catch your regulars when your food is already in front of them. A QR card in every bag, a perk that isn't a lower price, links from Google and Instagram, and one-tap reordering will do most of the work.
- 1
Put a QR card in every bag
Pack a printed card with a QR code to your ordering page in app orders and takeaway bags. Regulars see it every time.
- 2
Give a reason that isn't price
Offer no platform fee, a free side on the fifth order or priority pickup. These are perks, not cheaper menu prices.
- 3
Link it from Google
Add your ordering URL to your Google Business Profile, Instagram bio and WhatsApp Business profile.
- 4
Make reordering one tap
Saved addresses, UPI through Razorpay and order history mean a regular can check out in seconds.
- 5
Track the split
Compare direct orders with app orders every month. Adjust perks for the items that move.
What does the site need: Razorpay, a delivery radius and kitchen printing?
To get through a busy Friday night, a restaurant ordering website in India needs Razorpay payments with working webhooks, a delivery radius check and automatic kitchen printing. If any of those break, your own ordering will give you more headaches than the apps ever did.
Payments. Razorpay is the usual pick in India. It handles UPI, cards, wallets and net banking. Set up the webhooks properly. If you don't, you'll get WooCommerce orders sitting on "pending payment" when the customer has already paid. I go through the fix in WooCommerce Razorpay payments failing: how to fix it.
Delivery radius. Skip the radius check and sooner or later someone 12 km away orders food you can't get to them hot. For Whitfield Service Station I wrote a custom plugin that geocodes the customer's address with the Google Maps API, blocks anything outside the delivery zone and charges a delivery fee based on distance. It works the same in kilometres. The details are in how to limit WooCommerce delivery to a radius.
Kitchen printing. Your staff shouldn't be glued to a tablet. On the same build, every paid order prints a receipt in the kitchen through PrintNode. That's the main reason the store keeps taking orders while staff are busy with customers. Here's the full setup: how to print WooCommerce orders automatically in a kitchen.
Whitfield is a UK business, so it uses Stripe. For an Indian restaurant you'd swap in Razorpay and keep everything else.
Who delivers the order if you leave the apps?
You have three options: pickup only, an on-demand courier such as Borzo or Porter from about ₹27 a drop, or your own riders once direct volume is steady. It's the question that decides whether direct ordering pays.
- Pickup only to start. You keep the full saving on every order and there's no logistics to sort out. It's the cheapest way to test demand.
- On-demand couriers. Borzo or Porter send a rider per order. Prices start from ₹27 and go up with distance, which is one more reason to have a delivery radius.
- Your own riders. Worth it once direct orders are steady. ChefDesk suggests dividing a rider's full monthly cost by the deliveries they actually complete, then comparing that with a per-order courier rate.
Own website vs SaaS ordering vs WhatsApp: which fits?
Your own WooCommerce ordering site fits restaurants with steady repeat demand that want full control of customers and data. New restaurants with no regulars are fine on Zomato and Swiggy alone. A POS vendor's SaaS ordering page suits chains already on that POS, and WhatsApp ordering works for small menus.
| Option | Who owns the customer | Ongoing cost | Best for |
|---|---|---|---|
| Zomato and Swiggy only | The app | 18–25% commission + GST | New restaurants with no regulars yet |
| SaaS ordering page from your POS | You, inside the vendor's system | Monthly subscription | Chains already on that POS |
| WhatsApp ordering | You | Low per-message fees | Small menus and very loyal regulars |
| Own WooCommerce ordering site | You, fully | Hosting + 2% gateway fee | Restaurants with steady repeat demand that want control |
The SaaS page from your POS vendor is quick to switch on, but you're renting it. With your own WooCommerce site you own the code, the data and the SEO, and you can add your own rules like radius pricing or time slots. If Shopify is on your list too, read Shopify vs WooCommerce and compare running costs and how much custom logic you need before you decide.
What should you ask a developer before you hire one?
Ask to see a live ordering site they built with radius checks and kitchen printing. Ask how they handle Razorpay webhooks, and who owns the hosting, domain and code when the job's done. Here's the list I'd use:
- Have you built online ordering before, with radius checks and kitchen printing, and can I see it live?
- How do you handle Razorpay webhooks so paid orders never stay stuck as pending?
- Can I switch off ordering on busy nights, or set opening hours and prep times?
- How fast does the menu load on a mid-range Android phone over 4G?
- Who owns the hosting, domain and code at the end?
- What does monthly maintenance cover once the site is live?
Want to turn your idea into a feature list before you talk to anyone? The free Project Scope Generator gives you a structured scope in a couple of minutes.
What should you do next?
Pull last month's Zomato and Swiggy payout statements, count your repeat customers and put your own numbers into the break-even table above. If the maths works for you, have a look at my WooCommerce stores and online ordering service or book a 20-minute call. I'll tell you what a lean ordering site with Razorpay, a delivery radius and kitchen printing would need for your restaurant.
