Ask five studios what an app costs and you will get five ranges, all of them wide, none of them explained. That is not evasion so much as an honest reflection of how little anyone knows before discovery. But it leaves you unable to compare quotes, which is the actual problem.
Here is how the number is built.
The estimate is a function of four things
Scope, measured in decisions rather than screens. Twelve screens that show data are not the same work as four screens where users create it, approve it, and export it. Ask any estimate to be broken down by capability, not by page count.
Integration surface. Every external system your software must talk to adds work that is difficult to see from the outside: authentication, error handling, rate limits, and the days spent waiting for someone else’s sandbox credentials.
How well defined the requirements are. A specification that has been written down and agreed can be estimated within perhaps twenty percent. A conversation cannot be estimated at all, which is why the ranges you get before discovery are so wide.
Who is on the team. Senior engineers cost more per hour and less per project. This is not marketing; it is the most reliable pattern in the industry.
What reliably makes it more expensive
- Real-time features. Anything with live updates, presence, or collaborative editing is a different class of engineering.
- Payments and money movement. Correctness requirements are absolute and the edge cases are numerous.
- Multiple user roles with genuinely different permissions.
- Offline support that has to reconcile conflicting changes.
- Compliance requirements, especially when they arrive late.
- Migrating data from a system nobody fully understands any more.
What reliably makes it cheaper
- Cutting version one to the single thing the product must do.
- Deciding the brand and design direction before engineering starts.
- Having one person on your side who can make decisions without a committee.
- Accepting a well-understood technology choice over a novel one.
How to read an estimate
A number with no structure behind it is not an estimate, it is a guess presented confidently. What you want to see is the assumptions written down, the scope broken into pieces you recognise, the parts that are uncertain flagged as uncertain, and a statement of what would change the price.
If a studio will not put its assumptions in writing, the low quote is not a low quote. It is a change request waiting to happen.
What we do
We run discovery as a fixed-price engagement that produces a written specification and an estimate grounded in it. You own that document either way. If you take it elsewhere, it will still be useful, and if a competing quote comes in lower against the same specification, that is a real comparison rather than a guess against a guess.
If you are weighing up a project, tell us about it and we will give you a straight answer about what it involves.