Ask what custom software costs and the honest first answer is a shrug and a range: a focused internal tool might be a few thousand dollars; a platform the whole business runs on can be six figures. That range is real, and it is also useless to you. What you actually want to know is where your project lands inside it and why — so you can budget without either lowballing yourself into a broken build or overpaying for gold plating you will never use.
Here is how software is actually priced, what moves the number up and down, and how to get a figure you can rely on before a contract is signed.
Scope is the number; everything else is detail
The single largest driver of cost is how much the software has to do. Not how hard any one feature is — how many of them there are, and how they interact. A tool that does one thing for one kind of user is cheap to reason about and cheap to build. A system with several user roles, an approval flow, integrations with three outside services, and a reporting layer is not three times the work; it is more, because every part has to be correct in combination with every other part.
This is why the most valuable thing you can bring to a pricing conversation is not a budget — it is clarity about what the software must do and, just as importantly, what it must not. A precise scope is the difference between a real quote and a guess.
What pushes the number up
Integrations with other systems, especially ones with poorly documented or unstable APIs. Multiple user types with different permissions. Anything real-time. Handling money, which raises the correctness and security bar sharply. Compliance requirements. Custom design rather than a clean, systematic UI. And ambiguity — if the requirements are still forming, the price carries the cost of discovering them mid-build, which is the most expensive place to discover anything.
None of these are reasons not to build. They are simply the things that, when present, mean the number is higher — and knowing which ones apply to you explains most of the spread in any estimate you receive.
What keeps it down
A tight, well-understood scope. Reusing proven patterns instead of inventing everything. Buying the commodity parts — authentication, payments, email — rather than building them. Starting with the one workflow that matters most and expanding once it has proven its value, rather than commissioning the entire platform on day one.
The cheapest software is the software you did not have to build because an off-the-shelf tool genuinely fit. A good partner will tell you when that is the case, even though it means a smaller engagement.
The number nobody quotes: total cost of ownership
The build price is the smallest number in the room. Software has to be hosted, monitored, patched, and evolved as the business changes, for as long as it lives. A quote that ignores maintenance is not a real quote — it is a down payment on a surprise. Ask any prospective partner what ongoing care looks like and what it costs, and treat a vague answer as a red flag.
Budget for the life of the software, not just its birth. A build that is a little more expensive up front but designed to be cheap to maintain almost always wins over five years.
How to get a figure you can trust
Insist on a written scope before a price, and a price before a contract. A credible engagement starts with a discovery step — a working session that produces a written description of what is being built and clear success criteria — and prices against that, not against a hallway conversation. Fixed-price for well-defined work; a transparent weekly or monthly rate for genuinely exploratory work where the scope cannot yet be pinned down.
Be wary of a number produced instantly with no questions asked. Either it is padded to cover the unknowns, or it is about to become the first of several. The estimate you can trust is the one that came after someone understood what you actually need.
We scope every engagement in writing before quoting it, price fixed where the work is well-defined and transparently by the week where it isn't, and tell you plainly when buying off-the-shelf is the cheaper, honest answer. The clarity is the first deliverable — and it is the cheapest one.
- Build vs. buy: when custom software is actually worth itA straight framework for deciding whether to build custom software or buy off-the-shelf — and how to avoid the expensive mistakes on either side.
- How to choose a software development partnerHiring the wrong software firm is expensive and slow to discover. Here's what actually separates a partner worth keeping from one you'll be re-hiring around.