Skip to content

Planning your software budget · 5 min read

What changes the cost of a software project?

“It only needs a few screens.” That describes what people will see, but leaves out what they must be able to do. To understand a software estimate, follow one request through the business and ask which decisions, connections, and checks sit behind it.

Start here.

The work behind a feature matters as much as how it looks. Different access rules, business exceptions, existing systems, old records, and unresolved decisions can change what needs building and checking. Ask the supplier to explain those assumptions, identify the largest unknowns, and show which simpler choices would still finish the job.

  • Ask what must happen before and after each visible action; counting screens leaves that work unexplained.
  • Separate a known piece of work from a question the supplier still needs to investigate.
  • Reduce the first release by choosing fewer complete tasks, while keeping the checks and help those tasks need.

What has to happen behind a simple request?

Imagine a business asking for a button that lets customers change a booking. This is an invented example, not a client project or a price estimate.

The button is only the entrance. Can customers choose any available time, or must staff approve the change? How close to the appointment can they move it? What happens if another customer takes the same time while they are deciding?

Your team also needs to know what changed. The customer needs a clear result, and someone needs a way to resolve a change that did not finish. These are parts of the requested job even if most of them fit on the same screen.

Ask the supplier to walk through an ordinary change and an awkward one. The following planning examples show how a short request can leave different amounts of work to explain. They do not assign prices to those choices.

Hypothetical choices behind “let customers change a booking”. These are discussion prompts, not quoted features or cost rankings.
Possible first releaseWork to explainQuestion for your business
Customer sends a change requestStaff receive it, agree a time and confirm the resultCan staff handle these requests within the service you promise?
Customer chooses another available timeAvailability is checked and the booking is changed and confirmedWhich changes can happen without staff approval?
Different customers have different change rulesThe app identifies the applicable rule and handles exceptionsDo those differences matter for the first release?
The change must also reach an existing calendarBoth records need an agreed result and a way to resolve a failed updateWhat has been checked about the existing calendar?

Which decisions are still unanswered?

An estimate can contain both understood work and assumptions. Ask the supplier to point out the parts that depend on an answer they do not yet have. “We have not checked whether your current tool allows this” is useful information before you plan around it.

For the booking example, you may not yet know whether existing appointments can be changed through the other system. Or your own team may disagree about which customers should be allowed to move them. Those are different questions, with different people needed to answer them.

Ask for a bounded way to investigate an important unknown: what question will be answered, what work is authorised, what you will receive, and when you will decide whether to continue. A demonstration with sample records or a written rule may be the useful result.

If an assumption changes, ask which part of the estimate changes with it. You need an explanation tied to the new information, plus a chance to choose the next step before additional work begins.

What existing tools and records are part of the job?

List what the new app must read from or send to other tools. Also list what staff could continue handling themselves. “Connect our calendar” leaves more open than “show these appointments, let this person change them, and confirm the change in both places”.

Ask what has been checked about access, the available connection and failure handling. The supplier should explain what your business or the existing provider must supply. A familiar product name does not establish that the particular handoff is available.

Bringing old records across is another piece of work to discuss. Which records are needed at launch, who will resolve unclear details, and how will the transferred result be checked? Keep that one-time move separate from future updates between tools.

You can then consider alternatives with their business consequences visible. Staff might review change requests in the first release, for example. That still leaves staff work to plan and may not suit a business with a large number of requests. It is an option to assess, not an automatic saving.

How do the timetable and your team’s decisions affect the plan?

Tell the supplier which dates matter and why. An event, a contract ending, or a period when staff are available may shape the release. Ask what can realistically be completed and checked by that date, and which assumptions the answer depends on.

Make time for the decisions your business owns. Someone must approve the booking rules, try the working journey and respond when a question changes the agreed work. Name that person and a backup rather than assuming every review will happen immediately.

Ask when you will see the work and which choices need settling at each point. Trying a sample booking journey can expose a missing business rule before you ask the supplier to apply it across more of the app.

Keep the build price, your team’s preparation time and continuing service costs visible as separate commitments. Moving a task to your staff changes who does it; it does not make that task disappear.

How can you make the first release smaller without breaking its purpose?

Pick a significant item in the estimate and ask for alternatives. For our example, you could compare a staff-reviewed change request, a limited set of customer changes, and a broader booking service later. Ask what each option lets someone finish and what still happens outside the app.

Choose the version that fits the actual need and the work your team can handle. You might limit the first release to one kind of appointment or one customer group. Make those limits clear to the people who will use it.

Keep the essentials for that version: correct access, dependable saved information, understandable errors and a route to help. Removing a check that the chosen task relies on is different from deciding not to offer another task yet.

Finish with an updated description of the agreed work, remaining assumptions and continuing responsibilities. If the proposed budget still does not fit, discuss a narrower result, an existing product or delaying the build. The useful outcome is a commitment you understand, not a lower total with the missing work left unexplained.

Ask for the decisions behind the estimate.

You should be able to discuss the main choices without learning how to build software. For a significant item in the estimate, ask the supplier to fill in these four points with you.

The business result
What will a customer or member of staff be able to finish?
The work included
Which rules, connections, preparation and checks does that result require?
The unanswered question
What has not been established, how will it be checked, and when will you get a revised answer?
The smaller alternative
What could be delayed or handled another way, and what work would remain with your team?

Planning something with a similar workflow?

Discuss your project with us

A couple of common questions.

Can a supplier price an app from a list of screens?

A screen list can start a discussion, but ask what actions, rules, records and checks the price assumes. For anything unclear, agree how the supplier will investigate it and when you will receive a firmer description of the work.

Should I tell a supplier what I can spend?

Share your working limit and the result that matters most. Ask the supplier to explain what could fit, what would remain outside the project, and which assumptions need checking. A budget helps frame the choices; it does not establish that a particular result can be delivered for that amount.

Have something in mind?

Tell us what you want to build or improve. We’ll work out a useful first step together.

Discuss a project