In 2026, a custom e-commerce app costs roughly 10 to 20 lakh in India when it runs on the store you already have, about £8,900 to £17,700, and 20 to 60 lakh if it comes with a commerce platform of your own. A UK agency charges about £30,000 to £80,000 for the first and £60,000 to £150,000 and up for the second. Or you can skip the build: app builders that turn a Shopify store into an app cost $49 to $500 a month.
Those prices answer the question people type. The question that decides the money is different: will the app sell enough extra to pay for itself? An app does not find you new customers. It keeps the ones you have coming back, so whether it pays depends almost entirely on how many orders you already take. Below a certain volume, no app can earn back what it costs, however good it is.
In India there is a second answer, which a UK brand does not have. That is cash on delivery, and it is further down.
This guide is about the app. Which platform the store itself should run on, Shopify, headless or your own, is in Shopify against a custom store.
The three ways to get an app
| Route | What you get | India | UK agency |
|---|---|---|---|
| An app builder on your store | A branded app made from your Shopify catalogue, push notifications, limited custom screens | $49 to $500 a month, some take 2.5 percent of app sales | The same subscriptions, priced in dollars |
| A custom app on your existing store | Your own app reading the store's catalogue, cart and orders, with your checkout rules, push and analytics | 10 to 20 lakh, about £8,900 to £17,700 | £30,000 to £80,000 |
| An app and your own platform | Both the app and the commerce engine behind it, for rules no hosted platform handles | 20 to 60 lakh, about £17,700 to £53,100 | £60,000 to £150,000 and up |
Pound figures use 113 rupees to the pound, the rate across our cost guides, and Indian figures a blended 16,000 rupees a person-day. The middle row agrees with the e-commerce line in our app development cost guide, and the last row with the custom platform figure in our Shopify guide.
The first row is underrated. Builders such as Vajro, Tapcart and Plobal publish plans from about $99 to $500 a month, and the cheapest options start near $49. For a store testing whether its customers want an app at all, that is weeks and a few thousand rupees a month rather than months and lakhs. Watch the commission: a 2.5 percent cut of app sales stops being small the moment the app works.
Will an app pay for itself?
Here is the test, with numbers you can replace with your own.
A custom app on your store, as in the worked estimate below, costs about 11.2 lakh to build. Add a year of maintenance at 20 percent and roughly 8,000 rupees a month for push notifications, hosting and store accounts, and the first year costs about 14.4 lakh, or 1.2 lakh a month.
Say your average order is 1,500 rupees and you keep 30 percent of it after product and fulfilment: 450 rupees per delivered order. To cover 1.2 lakh a month, the app has to produce about 267 extra delivered orders a month, orders that would not otherwise have happened.
| Orders a month today | Uplift a custom app must create | Uplift an app builder at $249 a month must create |
|---|---|---|
| 1,000 | 27 percent | 4.6 percent |
| 3,000 | 8.9 percent | 1.5 percent |
| 5,000 | 5.3 percent | 0.9 percent |
| 10,000 | 2.7 percent | 0.5 percent |
| 20,000 | 1.3 percent | 0.2 percent |
Read the first row and the problem is obvious. A store taking 1,000 orders a month would need its app to lift sales by more than a quarter, which no app does. At 10,000 orders a month the same app needs under 3 percent, which a good one can manage from repeat customers alone.
So the rule of thumb: under about 3,000 orders a month, use an app builder or no app. Between 3,000 and 10,000, model it with your own numbers. Above 10,000, a custom app is usually worth building, provided a real share of your customers buy more than once. A store whose customers buy once a year will not see the uplift at any volume.
Cash on delivery, the lever only India has
Somewhere around 60 percent of Indian online orders are still cash on delivery, and logistics data puts the share of those that come back undelivered, return to origin, at roughly 24 to 27 percent. For prepaid orders the figure is about 3 percent. Every parcel that comes back costs you shipping both ways, packaging and handling, commonly 350 to 700 rupees once it is all added up, and earns nothing.
Put that into the same arithmetic. A cash on delivery order, with a quarter coming back at 350 rupees each, is worth about 250 rupees to you on average. The same order paid upfront is worth about 426. Each order you move from cash on delivery to prepaid is worth roughly 176 rupees.
That is where an app earns its keep in India. Saved UPI details, one-tap repeat payment, prepaid-only offers and cash on delivery limits by pincode or order value all push orders towards prepaid, and an app does them more easily than a website. Moving about 680 orders a month from cash on delivery to prepaid would pay for the custom app on its own, before counting a single extra sale.
The honest caveat: a good website checkout can do much of this too. One-click checkouts that verify addresses and nudge towards UPI exist for web stores, and they cost less than an app. Fix the checkout first. Build the app when repeat customers and prepaid share are both worth more than the checkout alone can deliver.
A UK store has no equivalent. Cards and wallets are already prepaid, so a UK app has to pay for itself on repeat sales alone.
Where the money goes: a worked estimate
A custom app for iOS and Android, built once in Flutter or React Native, reading from an existing Shopify store.
| Work package | Person-days |
|---|---|
| Discovery: which customers the app is for, what it must do that the website does not, and the number that will judge it | 3 |
| Design: home, catalogue, product, cart, checkout, account, orders | 10 |
| The app shell and navigation, one codebase for both platforms | 5 |
| Catalogue, search and filters from the store | 8 |
| Cart and checkout: UPI first, saved methods, cash on delivery rules by pincode and order value | 10 |
| Accounts, order history, courier tracking, return requests | 8 |
| Push notifications with consent: back in stock, price drops, abandoned carts | 6 |
| Analytics: which orders came from the app, and their prepaid share | 4 |
| Speed and offline caching for slow networks | 4 |
| Testing on real low and mid-range Android phones, and both stores' reviews | 8 |
| Launch and handover | 4 |
| Total | 70 |
That is about 11.2 lakh, roughly £9,900 or $13,300, over ten to twelve weeks. Note the size of the checkout line. Ten days for cash on delivery rules, UPI and saved payment methods is more than the catalogue gets, and in India it is the part that earns the money back.
For a UK brand
The same test, run for a UK store with a £60 average order and 30 percent kept, gives £18 per order.
| Built by | First year cost | Extra orders a month to pay back |
|---|---|---|
| A UK agency, £50,000 build | About £68,400 | About 317 |
| An Indian team, the worked estimate | About £12,700 | About 59 |
That is the UK case for building in India in one table. The app is the same; the order volume needed to justify it falls to about a fifth. The fully loaded comparison, including what it takes to work across time zones, is in our guide to hiring developers in India.
Two things stay UK-side whoever builds it: UK GDPR, which means an offshore team works on test data and reaches customer data only through controlled access, and the payment setup, where Apple Pay, Google Pay and pay later options replace UPI and cash on delivery in the checkout design.
The costs that are not in the quote
| Cost | Typical size | When it surprises you |
|---|---|---|
| App store accounts | Apple $99 a year, Google $25 once | Small, but review cycles cost days |
| Builder commission | Up to 2.5 percent of app sales on some plans | The month the app starts working |
| Push notifications and hosting | A few thousand rupees a month, more at volume | Once you send campaigns to the whole list |
| Maintenance | 15 to 25 percent of the build a year, as in our app maintenance guide | Every OS release |
| Getting installs | Marketing to existing customers, discounts for first app orders | Always; an app nobody installs earns nothing |
| Keeping the app and website in step | A share of every feature, built twice | When the website gets a feature the app does not |
GST at 18 percent applies to development invoices for Indian clients and is claimable as input credit if you are registered. Services exported to UK and other overseas clients carry no Indian GST.
The six things that move the price
- Whether you already have a store. An app reading from Shopify is half the work of one that needs its own platform.
- Checkout rules. Cash on delivery limits, prepaid offers, partial payments and pincode rules are where Indian apps differ most from templates.
- One codebase or two. Flutter or React Native for both platforms is cheaper than separate native apps, and the trade-offs are in Flutter against React Native against native.
- Personalisation. Recommendations and personal offers add data work and testing, and are worth it only at volume.
- Integrations. Couriers, loyalty, reviews and a CRM each add a few days and a system that can change under you.
- How custom the design is. A design system built for the app is a premium choice; a clean, standard design gets most stores most of the value.
How to read an e-commerce app quote
- Ask what the app does that the website does not. If the answer is "everything the website does, in an app", it will be hard to pay back.
- Ask how cash on delivery is handled. Rules by pincode, order value and customer history should be in scope, not a later phase.
- Ask which numbers the app will report. App orders, repeat rate and prepaid share, from launch.
- Ask who keeps the app and website in step when either changes.
- Ask for the year one total, not only the build.
Our wider checklist for choosing who builds it is in how to hire a development company in India.
Questions people ask
How much does an e-commerce app cost in India?
Ten to 20 lakh for a custom app running on your existing store, 20 to 60 lakh for an app with a commerce platform of your own, or $49 to $500 a month for an app builder on Shopify. The worked estimate above, a Flutter or React Native app on a Shopify store, comes to 70 person-days and about 11.2 lakh.
How much does an e-commerce app cost in the UK?
UK agencies publish roughly £30,000 to £80,000 for a mid-sized shopping app and £60,000 to £150,000 for a retail app with deeper integration. The same app built by an Indian team costs about £8,900 to £17,700, and app builder subscriptions cost the same in either country.
Does my store need an app?
Probably not under about 3,000 orders a month. A custom app costing about 14.4 lakh in its first year needs around 267 extra orders a month to pay back at 450 rupees kept per order, which is a 27 percent lift for a store taking 1,000 orders. Above 10,000 orders a month the required lift falls under 3 percent, and an app usually pays.
Is an app builder good enough?
For most stores starting out, yes. Builders give you a branded app, push notifications and your catalogue in weeks. You outgrow them when you need checkout rules, screens or integrations they cannot do, or when their commission on app sales exceeds what a custom app would cost to run.
How does cash on delivery affect an app?
It makes prepaid orders much more valuable. Around a quarter of cash on delivery orders come back undelivered, against about 3 percent of prepaid ones, so each order moved to prepaid is worth roughly 176 rupees on a 1,500 rupee order. An app's saved UPI details and prepaid offers make that shift easier, though a good website checkout should come first.
How long does an e-commerce app take?
Ten to twelve weeks for a custom app on an existing store, plus a week or two for app store reviews. An app builder can be live in days. An app with its own platform takes five months or more.
Should the app be native or cross-platform?
Cross-platform, with Flutter or React Native, for almost every store. One codebase serves both iOS and Android, costs less to build and maintain, and handles everything a shopping app needs. Native apps are worth it only for unusual demands such as heavy augmented reality.
What does an app cost to run each month?
A few thousand to a few tens of thousands of rupees for hosting, push notifications and store accounts, plus 15 to 25 percent of the build a year in maintenance. On an app builder, the subscription and any commission on app sales replace most of that.
Getting a real number
Ranges end where your orders begin. Tell us your monthly orders, average order value and cash on delivery share, and we will tell you whether an app pays for itself, which route fits, and what it costs. Sometimes the answer is a better checkout, not an app. What we build for online stores is on our e-commerce development page, and custom platforms on our e-commerce platforms page.