A business case is how you turn an idea into a decision. Done well, it gives the reader a clear go or no-go and the numbers to defend it. This is the step-by-step method we use for investment cases, new initiatives, and acquisitions. The structure is the same whether you are asking for US$50,000 for a pilot or US$50 million for an acquisition. The discipline is what earns the approval.

Step 1: Start with the decision, not the idea

Write the decision you want made in one sentence: "We should invest US$2.4 million to automate the warehouse, returning an IRR above 18% within four years." If you cannot state the decision crisply, you are not ready to write the case. Everything else in the document serves this decision. Pin the sentence to the top of the page and keep returning to it as you draft. Any section that does not help a reader say yes or no to that sentence is a candidate for cutting.

Step 2: Make the strategic case ("why now")

Before the numbers, argue why this matters now. Name the market shift, the competitive window, the regulatory change, or the customer pain that makes this quarter the right quarter. In the warehouse example, the trigger might be that two competitors have already automated and cut their fulfilment cost per order by 12%, while your manual cost has risen 4% year on year with wage inflation. That combination is the "why now". Tie the case to a real strategic rationale, not "it would be nice". A senior reader should be able to repeat your why-now argument back to you after reading the first page.

Step 3: Lay out the options, not just the plan

A one-option business case is a wish. Good cases compare two to four options: do nothing, invest small, invest large, partner, or rent. Quantify each at the same level of detail so the comparison is fair. "Do nothing" is always an option, even when the answer is that doing nothing loses value faster. In our example:

Option A (do nothing): continue manual fulfilment, cost per order stays at US$4.10 and climbs with wages.
Option B (partial automation): invest US$1.1 million in one automated packing line, cut cost per order to US$3.20 across 60% of volume.
Option C (full automation): invest US$2.4 million in an integrated line and warehouse software, cut cost per order to US$2.60 across 90% of volume.
Option D (partner): outsource fulfilment to a 3PL, trading fixed investment for a per-order fee of US$3.05.

Now the recommendation writes itself. Full automation wins on cost per order but carries the most execution risk, and the case should say why the extra risk is worth the extra US$0.60 per order versus partial automation.

Step 4: Build the financial case on explicit assumptions

For each option, lay out the key financial lines: the investment required, the revenue build, the cost base, and the returns. Every number must trace back to an assumption you can state. Write the assumption in the cell next to the number, not buried in an appendix. If a number is not yet validated, mark it "[to be validated]" rather than inventing a figure.

For the full automation option the model might look like this:

Capital investment: US$2.4 million (two automated packing lines at US$950,000 each, plus US$500,000 of warehouse software and integration).
Annual volume: 1.8 million orders per year, flat in the base case.
Cost per order today: US$4.10, falling to US$2.60 with automation.
Gross annual saving: 1.8 million x (US$4.10 - US$2.60) = US$2.7 million.
Running costs of the system: US$900,000 per year for maintenance, energy and retained labour.
Net annual benefit: US$1.8 million before tax, before depreciation.
Payback on the US$2.4 million: roughly 16 months if benefits ramp in the first year.

Writing it this way means anyone can challenge any single number without the whole model falling apart. If volume is really 1.4 million orders, the net benefit drops to about US$1.2 million and payback stretches past two years. That is not a flaw; that is the model doing its job.

Step 5: Show the returns honestly

Present ROI, IRR and payback, and be honest about the range. A case that only shows the upside is a red flag to an experienced reader. Show the base case and the downside, and say what would have to go right to hit each.

Base case: net benefit US$1.8 million per year, IRR of 23%, payback in 16 months.
Downside case: volume at 1.4 million orders, implementation slips six months, running costs overrun by 20%. Net benefit falls to around US$1.0 million, IRR drops to 11%, payback pushes past two years.
Upside case: volume grows 8% per year and the line hits rated capacity in month nine. IRR climbs above 30%.

The honest framing is the selling point. A sponsor who sees the downside already modelled trusts the case, because it reads like analysis rather than advocacy. Add a one-line sensitivity statement, for example: "Every US$100,000 of capital cost overrun cuts IRR by roughly 1.5 percentage points."

Step 6: Name the risks and the mitigations

List the top risks and, for each, what you will do about it. "Integration risk, mitigated with a dedicated PMO and a two-week dry run on the existing packing line" is a plan; "integration risk" alone is not. This is what separates a case that gets funded from one that gets questioned. For the automation case:

Implementation overrun: mitigated with a fixed-price vendor contract and a penalty clause.
Volume shortfall: mitigated by commissioning one line first and only approving the second on a volume trigger.
Labour disruption: mitigated with a retraining programme and a six-month no-redundancy commitment.
Technology becoming obsolete: mitigated with a one-year warranty and a software-agnostic integration layer.

Assign an owner to each risk. Risks without owners are wishes.

Step 7: Make a clear recommendation

End with an unambiguous recommendation: go, no-go, or invest with conditions. If the answer is conditional, state the conditions explicitly. For the automation case: "Recommend go on full automation, conditional on the fixed-price contract being signed at or below US$2.4 million and on the volume trigger for the second line being approved in the quarter it is reached." A business case that waffles at the end undoes all the work that came before it.

Step 8: Add next steps

Close with three to five concrete next actions: who does what, and by when. "CFO signs the vendor contract by 30 June; plant manager completes the commissioning plan by 15 July; HR publishes the retraining schedule by 31 July." This turns the approval into momentum and gives the follow-up review something to check against.

Common mistakes to avoid

Building the case around the answer you want instead of the decision to be made.
Hiding the downside, which destroys credibility with exactly the reader you need on your side.
Using a single growth-rate model that cannot be stress-tested line by line.
Submitting a case with no named owners or dates, which stalls the moment it is approved.
Comparing options at different levels of detail, which makes the simple option look cheap and the full option look risky.

Speed it up

Scaffold the structure with the Business Case generator. It builds the executive summary, options, financial case and risks, then export as a deck or Word report.

How to write a business case (step by step)