Disclosure up front: I work at ZoopCoder, an Indian web development agency. We sell website builds, so read this knowing what I sell. Every rupee figure below is either our own published price or clearly labelled as a market rate you should verify yourself.
Indian clients ask for a per-page rate constantly. "We need 12 pages, what's your rate per page?" It is a completely reasonable question and it has no useful answer, and I want to show the arithmetic rather than just assert that.
The same page count, three very different numbers
Here is an 8-page site, built three ways, at our published prices:
| Build | Price (INR) | Per page |
|---|---|---|
| 8-page brochure site (static content, contact form) | 25,000 – 50,000 | 3,100 – 6,250 |
| 8-page WordPress/CMS (client edits content) | 35,000 – 75,000 | 4,400 – 9,400 |
| 8-screen custom web app (user logins) | 1,00,000+ | 12,500+ |
Identical page count. Roughly a 4x spread on the per-page number. A rate quoted before anyone knows what the pages do is going to be wrong in one direction or the other — and in my experience it is usually wrong in the direction that ends in an awkward conversation three weeks in.
The cliff is not pages. It is state.
The reason the per-page frame breaks is that it measures the wrong unit.
A brochure page is a pure function of nothing. Same bytes for every visitor. You can cache it at the edge forever and the server never thinks. Adding the 9th, 10th and 20th brochure page is close to linear and genuinely cheap — a few thousand rupees each, mostly content and layout work.
The moment you add a login, every page becomes a function of who is asking. And that one change drags in a pile of work that is invisible on every wireframe:
- accounts, sessions, password reset, email verification
- authorization — not "is this user logged in" but "is this user allowed to see this specific record"
- a database of user-owned data, with migrations
- the negative case: never leaking user A's data to user B
- and the tests for all of the above, because this is the layer where a bug is a breach, not a typo
None of that shows up as a page. None of it is design. All of it has to be built, and more importantly, tested. That is where the money between 75,000 and 1,00,000+ actually goes.
So the price cliff in the Indian market sits right about there — and it is a cliff of statefulness, not of page count. This is why "we'll just add a small login area" is the single most expensive sentence a client can say in a kickoff call. It is not a small addition. It is a different category of software.
The practical version, including the part that costs us money
If your site has no login, do not buy a web app. If nobody can name a single thing on the site that requires a user to sign in, you are buying a brochure site, and a brochure site in India costs 25,000 – 75,000 depending on whether you want to edit the content yourself. We see people quoted 1,50,000+ for exactly this. If that is your project, you do not need us at that price — and honestly you may not need an agency at all.
If your site does have logins, be suspicious of a 40,000 quote. Either the scope has not been understood and the number will move, or someone is going to skip the authorization work, which is the part you cannot see and cannot easily check on delivery.
The costs the per-page number hides entirely
Whatever you build, the build price is not the annual price:
| Item | Cost |
|---|---|
| Domain | 800 – 1,500 / year |
| Hosting (most business sites) | 3,000 – 15,000 / year |
| Maintenance plan (optional) | 3,000 – 15,000 / month |
That last row is genuinely optional and I want to be clear about it, because we sell those plans. A small brochure site that nobody logs into often does not need a monthly plan at all. A stateful app with user data and a payment flow does — not for the "content updates" bullet, but because dependency and security patches on a system holding user records are not something you skip.
Full by-type cost breakdown, with the tables: https://zoopcoder.com/website-development-cost-in-india.php
If you price web work in India differently — particularly if you have found a per-page model that holds up on stateful builds — I would genuinely like to hear it, because I have not.
Top comments (0)