Software Delivery · How-to

How to scope a custom software project so it lands on time and on budget

Most software projects that run late were scoped badly, not built badly. Here's how to define a feature or application clearly enough to get an accurate estimate, a realistic timeline and a result your team actually uses.

By Max Chuchmai3 min read

On this page
  1. Start with the business outcome
  2. Describe user journeys, not screens
  3. Split launch from later
  4. List every system it touches
  5. Agree what “done” means
  6. Choose the right engagement model
  7. Questions to prepare before the first meeting

When a software project runs over budget, the cause is rarely the code itself. Far more often, the problem started before any code was written: the scope was vague, assumptions were never written down, and different people imagined different products.

Whether you are planning a new internal application, a customer portal or a new feature in an existing product, the quality of the scope decides the quality of the estimate. This guide explains how we scope projects with clients, and how you can prepare.

Start with the business outcome

Before listing features, agree in one or two sentences what the software must change for the business. For example:

  • “Account managers can see every client’s orders, invoices and support tickets in one place, without logging into four systems.”
  • “Partners can register deals themselves, and our sales team is notified within minutes.”
  • “Customers can book and reschedule installations online instead of by phone.”

This outcome becomes the test for every feature request that follows. If a feature doesn’t serve it, it probably belongs in a later phase.

Describe user journeys, not screens

The most useful scope documents describe what people need to do, step by step, rather than what the screens look like. For each type of user, write the main journeys:

  1. Who is the user? (for example, a customer’s finance manager)
  2. What are they trying to achieve? (download last quarter’s invoices)
  3. What do they start from, and what do they need to see or do along the way?
  4. What happens when something goes wrong? (the invoice is missing, the payment failed)

Journeys reveal requirements that feature lists miss: permissions, notifications, error handling and edge cases. They also make it easy for designers and engineers to propose the simplest way to deliver them.

Split launch from later

Every project has more ideas than budget. Make the split explicit:

  • Must have for launch: without these, the software doesn’t deliver the outcome.
  • Should have soon after: valuable, but the business can go live without them.
  • Could have later: good ideas to revisit once people use the software.

Writing this down protects the timeline. When a new request appears mid-project, the conversation becomes “which of these does it replace?” rather than “can we squeeze it in?”.

List every system it touches

Integrations are the most common reason estimates go wrong. Early in scoping, list every system the software must read from or write to: CRM, ERP, billing, identity and login, email, payment providers, data warehouse. For each, note:

  • What data flows, and in which direction.
  • How often: real time, hourly, daily.
  • Whether the system has a usable API, and who can grant access.
  • Who owns the data if the two systems disagree.

Our guide to integrating business systems goes into more detail.

Agree what “done” means

“Done” means different things to different people. Agree up front that a piece of work is only done when it is:

  • Built and reviewed.
  • Tested, including the error cases from the user journeys.
  • Secure, with appropriate access control and data protection.
  • Documented well enough for someone else to maintain it.
  • Accepted by the business owner against the agreed journey.

This definition is the difference between software that works in a demo and software that works on a busy Monday morning.

Choose the right engagement model

How you contract the work should match how certain the scope is.

  • Fixed scope and price works well when journeys, integrations and priorities are clear. It gives budget certainty.
  • Time and materials suits work where the right answer will only emerge as you build, such as a new product or an experimental feature.
  • A paid discovery phase first is often the best of both: a short, fixed-price engagement that produces the scope, designs and estimate, after which the build can be fixed-price.

Questions to prepare before the first meeting

  • What business outcome must this software deliver, and how will we measure it?
  • Who are the users, and what are their three most important journeys?
  • Which systems does it need to connect to, and who controls access to them?
  • What is the deadline, and what drives it?
  • Who on our side makes decisions and accepts the work?

Arriving with even rough answers to these questions shortens the scoping phase and makes the estimate far more reliable.


We scope, design and build custom applications with a written scope and a fixed price before development starts. See our custom software services or book a call.

Written by

Max Chuchmai

Co-Founder & CTO, CoderStein. System architecture, AI automation, and GTM engineering.

Follow on LinkedIn

Discuss your software, AI and GTM engineering needs.

30-minute consultation · Clear technical next steps within one business day

Book a Call