Custom software is priced by scope, not from a price list, because "custom software" covers everything from a single internal form that replaces a spreadsheet to a multi-role platform handling payments and inventory. The honest answer to what it costs is that it depends on four things: how many distinct workflows it supports, how many systems it has to talk to, how many types of user it serves, and how long it needs to keep running.
This post explains each of those, describes the tiers most business projects fall into, and gives you a way to compare proposals that does not reduce to picking the smallest number. We do not publish figures, and the last section explains why.
The four factors that move the number
- Number of distinct workflows. Not features — workflows. "Staff submit a leave request, a manager approves it, payroll sees the result" is one workflow with three steps. Counting workflows rather than features is the fastest way to size a project honestly, because features multiply while workflows stay countable.
- Integrations. Every external system — accounting software, a payment gateway, a courier API, WhatsApp, an existing database — is a separate contract with someone else’s rules and failure modes. Integrations are consistently the most underestimated line in any quote.
- User roles and permissions. One kind of user is straightforward. Admin, manager, staff and customer, each seeing different data with different rights, is a different class of problem. Permissions touch every screen and every query.
- Expected lifespan. Software meant to run for five years is built differently from a tool for one campaign: more tests, clearer structure, better documentation. Both are legitimate; they cost different amounts and you should say which you want.
The tiers most business projects fall into
- Internal tool. One team, one workflow, replacing a spreadsheet or a manual process. Small scope and often the highest return per rupee spent, because it removes work someone is currently doing by hand every day.
- Business system. Several connected workflows, a few user roles, an admin view and reporting. Inventory, bookings, job tracking, client management. This is where most small and mid-sized business projects genuinely sit.
- Customer-facing platform. Your customers use it directly. Accounts, payments, notifications, support for people who did not read a manual and never will. The step up in cost is mostly about the edge cases and the polish that external users require.
- Multi-tenant product. You are selling the software itself to multiple organisations. A different business model, and effectively product development rather than a project.
Why proposals for the same brief vary so widely
A wide spread between quotes is usually a spread in assumptions, not in greed. The gaps are almost always in the same five places.
- Discovery. Some quotes include mapping the workflow properly before building. Some assume your brief is already the specification. The second is cheaper right up until the first change request.
- Data migration. Getting years of existing records out of spreadsheets or an old system, cleaned and imported, is frequently as much work as a feature. It is often missing entirely from a low quote.
- Testing. Automated tests cost time up front and save it every month thereafter. A quote without them is smaller and more fragile.
- Hosting and deployment. Who sets up the server, the database, the backups and the monitoring, and who pays for them ongoing.
- Training and handover. Software nobody has been taught to use does not get used. This is a real line item and it belongs in the proposal.
Fixed price or time and materials?
A fixed price transfers risk to the studio, so it includes a buffer and requires the scope to be locked. It suits well-defined projects and clients who need budget certainty. Time and materials is cheaper when scope genuinely cannot be known up front, but it requires trust and active management on your side.
For most business software, the sensible arrangement is a paid discovery phase that produces a specification, followed by a fixed quote against that specification. You pay a small amount to remove the uncertainty, then buy the build with a number that can be relied on. We work this way because open-ended budgets tend to end badly for both sides.
How to compare two proposals properly
- Ask each to list what is explicitly out of scope. The answer is more informative than the feature list.
- Ask who owns the code, the data and the hosting accounts at the end. If the answer is not "you", the price is not the price.
- Ask what happens after launch, and what it costs. Support arrangements vary more than build prices do.
- Ask what they would cut if the budget were 30% lower. A good answer shows they understand which parts carry the value.
- Ask to see something comparable they have built and, ideally, to speak to that client.
Why we do not publish a price list
Any figure broad enough to be true for every project would be too broad to help you budget, and a narrower one would simply be wrong for most readers. Worse, a published number steers the conversation toward fitting a build to a price rather than scoping the software the business actually needs. We quote per project after understanding the workflows, and the figure is fixed in writing before work begins.
If you are trying to work out whether your problem warrants custom software at all, our software development service sets out how we scope a build, what the architecture stage produces, and what you receive at handover.
Frequently asked questions
Is custom software cheaper than a monthly SaaS subscription?
Not initially — custom software is a capital cost where SaaS is an operating cost. It becomes cheaper over time when subscription fees scale with your headcount or usage, or when the tool nearly fits but forces expensive manual workarounds. The crossover often arrives sooner than people expect once per-seat pricing grows, but it depends entirely on your numbers.
How long does a custom software project take?
A focused internal tool is usually a matter of weeks. A business system with several workflows and roles runs longer and should be broken into milestones so something usable ships early rather than everything arriving at the end. Any credible proposal commits to a milestone timeline before work begins.
What is a discovery phase and do I have to pay for it?
Discovery is the work of mapping your actual workflows, data and integrations, and turning them into a specification that can be quoted accurately. It is normally paid, because it is real work that produces a real deliverable — and the specification is yours to take to other studios if you want competing quotes against it.
What if my requirements change halfway through?
They usually do, and a good process expects it. Changes within the agreed scope are absorbed; changes that add scope are quoted as a variation before they are built, so you decide whether each one is worth its cost. What you should avoid is any arrangement where changes are silently absorbed, because that cost reappears as rushed work or a missed deadline.
Who owns the software when it is finished?
You should own the source code, the data and the hosting accounts outright, and it should say so in the contract before work starts. At Avenix Studio the repository, accounts and documentation are handed over at the end of every project, with nothing locked to a proprietary platform or to us.
Want a site that puts this into practice?
Book a call