How to Write a Business Case: Steps, Examples and Checklist

When you need this document

Write a business case when an opportunity has passed early validation and now needs funding, headcount or sustained organisational commitment — typically during the Validate phase, before significant delivery effort begins. It is also the right instrument when leaders must choose between competing investments. If you are still establishing whether the opportunity is real, stay with an opportunity assessment; the business case assumes that question has been answered.

Gather these first

• The validated opportunity or problem evidence the case rests on — ideally an opportunity assessment or validation findings
• Cost inputs from the people who will do the work: engineering, design, operations, plus ongoing run costs
• Commercial assumptions you can defend — pricing, demand, adoption, churn — with their sources
• The genuine alternatives, including a costed do-nothing baseline
• Finance's view on how returns should be modelled in your organisation
• The executive sponsor whose endorsement the funding decision will need

Step by step

1. Start with the decision, not the document

   Before writing anything, be able to say in one sentence what you are asking for: how much, over what period, to achieve what return, decided by whom. Every section you write exists to support that single sentence. Cases drift when authors start with the background and hope a decision emerges; strong ones are engineered backwards from the ask.

   What good looks like: You can state the ask — amount, duration, expected return, decision-maker — in one sentence before drafting begins.

2. Frame the problem in commercial terms

   Describe the problem or opportunity in the language of the people funding it: revenue at stake, cost being incurred, customers being lost, risk accumulating. Customer pain matters, but a funding audience needs to see the mechanism by which that pain becomes money, growth or exposure. Anchor the framing in evidence — what is happening now, and what it costs to leave it alone.

   What good looks like: A finance reviewer can see the commercial mechanism — how the problem converts to money, growth or risk — without a product person explaining it.

3. Show strategic fit as a choice, not a slogan

   Explain where this investment sits against the organisation's stated direction and current portfolio. The persuasive version is specific: which strategic priority it advances, what it depends on, and what it competes with for the same people and money. Leaders approve cases partly on opportunity cost, so name what this displaces — pretending it displaces nothing reads as naive.

   What good looks like: The fit section names the priority served and is honest about what competes for the same resources.

4. Cost it honestly, including the unglamorous parts

   Build the full cost picture with the people who will deliver it: build effort, ongoing run and support costs, tooling, training, migration, and the cost of maintaining it in year two and beyond. Under-costing is the most common way cases fail in retrospect. Present costs with the same honesty you would want if you were the one approving — ranges where uncertain, and a clear note on what is excluded.

   What good looks like: The cost section includes run, support and maintenance beyond launch, and states its confidence level.

5. Make every commercial assumption visible and testable

   The return side of your case rests on assumptions about pricing, demand, adoption and behaviour. List them individually, source each one, and flag which are strong and which are hopeful. A reviewer should be able to change one assumption and see how the case moves. Burying assumptions inside a spreadsheet total is how cases pass review and then quietly fail in the market.

   What good looks like: Each assumption is stated, sourced and rated, and the case shows which single assumption it is most sensitive to.

6. Compare real options, and cost the do-nothing path

   Present the alternatives you genuinely considered — a smaller version, a partner or buy option, deferring a year, doing nothing — and say why the recommended path wins. The do-nothing option deserves real numbers: what continuing as-is costs in lost revenue, mounting risk or team drag. Straw-man alternatives are transparent and cost you credibility on everything else.

   What good looks like: Every option presented is one a reasonable leader might actually choose, and doing nothing carries a cost figure.

7. Confront the risks before your reviewers do

   List the material risks — delivery, market, commercial, operational — with a named mitigation or an honest acknowledgement that one does not exist. The goal is not to look safe; it is to show you have seen what could go wrong and have thought about it first. A case whose risk section is thinner than its benefits section signals advocacy, and experienced reviewers read it that way.

   What good looks like: The risks section contains at least one risk serious enough that a reviewer would have raised it — with your answer already attached.

8. Write the executive summary last, and make it decision-ready

   Once the body is settled, distil it: the problem, the ask, the expected return, the key risk, the recommendation. Many readers will decide from this half-page alone, so it must stand on its own and match the body exactly. Close the document with the specific decision requested, success measures for after approval, and the named owner — so approval converts directly into action.

   What good looks like: A leader reading only the summary reaches the same conclusion as one who read every section.

Common mistakes

• Benefits are precise while costs are vague — a £3.2m return against 'approximately two squads for a while'. — Match the rigour on both sides. If benefits get a model, costs get a model, including run and maintenance. Asymmetric precision is the fastest way to lose a finance reviewer's trust.
• Assumptions are buried in the numbers, so the case cannot be challenged — only accepted or rejected whole. — Pull every commercial assumption into its own labelled list with sources and confidence. A challengeable case is a stronger case; reviewers who can probe individual assumptions approve with more conviction.
• The options section is theatre — two straw men flanking the author's preferred path. — Include at least one alternative you would genuinely accept if the recommendation were rejected, and give doing nothing an honest cost. Reviewers have seen a thousand rigged comparisons.
• No one can tell what happens if the case is approved — the money is granted and momentum stalls. — End with named owners, first actions, success measures and a review date. Approval should trigger motion, not a second round of planning.
• The executive summary oversells relative to the body, and reviewers who read both notice the gap. — Write the summary last, from the finished body, and have someone check the two tell the same story. Any daylight between them reads as spin.

Before you call it done

• The ask is one sentence: amount, duration, expected return, decision-maker
• The problem is framed in commercial terms with evidence of what inaction costs
• Costs include build, run, support and maintenance, with confidence stated
• Every commercial assumption is listed, sourced and rated for confidence
• At least one alternative is genuinely viable, and do-nothing carries a number
• Each material risk has a mitigation or an honest admission that none exists
• The executive summary matches the body and could carry the decision alone

Frequently asked questions