What a business case is for
Most requests for marketing money fail in one of two ways. Some are too vague to approve: "more budget for video" cannot be costed, measured or stopped. Others promise too much: a single confident forecast that finance knows is uncertain, and that the requester will later be held to. A business case avoids both by being specific about the decision and honest about the uncertainty.
It is different from the neighbouring documents. The budgeting methods guide sets the total marketing budget, and the tutorial on building a marketing budget turns that total into a document finance can approve and track. The budget reallocation memo moves money already approved. A business case asks for something new, one investment at a time, usually outside the annual cycle or as a distinct line within it.
Finance fluency is part of the job. The site's marketing finance hub notes that professional bodies such as the Chartered Institute of Marketing treat financial literacy as part of marketing management rather than an optional extra. In practice that means presenting the case in the terms finance uses: profit rather than revenue, all costs rather than media alone, and evidence of cause rather than correlation.
The investment type changes what the case has to show.
| Investment | Costs most often missed | How the result is usually proved | Natural first stage |
|---|---|---|---|
| A new channel | Creative production, the learning period, the team's time to run it | A bounded test, ideally with a holdout or regional comparison | A trial period with a fixed budget and a decision point |
| A brand campaign | Production, the long tail of effect beyond the campaign, research | Branded search, direct traffic, new-customer share, a regional comparison where possible | One region or one flight before the full plan |
| A tool | Implementation, integration, training, migration, the renewal price | Time saved, errors avoided, work made possible, against a before measure | A trial or proof of concept with the business's own data |
| A hire | Recruitment, onboarding, management time, the months before full productivity | The work the role makes possible and what it replaces, such as agency fees | A defined first-year scope with review points |
Before you start
- The business's own numbers. Contribution margin per sale, conversion rates through the funnel, past campaign results, average order or contract value and repeat rates. The case's ranges are built from these.
- Finance's view of costs. Day rates or salary costs for internal time, how the business treats one-off and recurring spend, and the payback period it usually expects.
- The decision-maker and the timing. Who approves an investment of this size, when they decide and what format they expect.
- The alternatives. At least one other use of the same money, and the option of doing nothing.
- A conversation with finance early. A partner in finance who has seen the method before it is presented is far more useful than one meeting it for the first time in the approval meeting. If the business already distrusts marketing's numbers, the diagnosis of finance distrusting marketing numbers identifies the usual causes, and fixing them helps every future case.
The steps
State the decision and the alternatives
Write the decision in one sentence that can be approved, reduced or declined: what money, for what, over what period, with what first commitment. "Approve a first-quarter budget to produce project videos and run them as paid video ads in two of our four postcode areas, with a decision at the end of the quarter on whether to continue" can be decided; "invest in video" cannot. Then list the alternatives: the same money in an existing channel, a smaller version of the same idea, and doing nothing. Put a value on doing nothing where you can, such as demand going to a competitor, a contract that will be harder to renegotiate later or work the team cannot do without the hire. The comparison is what makes the case a decision rather than a request. The value of the best of those alternatives is the opportunity cost of the investment, and it is the bar the request has to clear.
Cost it completely
List every cost the investment brings, at full cost, separating one-off from recurring. Media; production of creative, video or content; agency and freelance fees; tools and their implementation; recruitment and onboarding for a hire; and the time of the people inside the business who will run, review and manage it. Internal time is the cost most often left out and the one finance most often notices; cost it at what the business pays for that time, even where it needs no new budget. For a tool, the guide to choosing and evaluating tools covers total cost and contract terms; for a hire, the guide to hiring and developing marketers covers scoping the role before advertising it. Present costs by month so the reader sees when money leaves the business.
Estimate the upside as ranges
Build the expected result from the business's own data, not from industry averages or a vendor's case studies. Work through the chain the investment depends on: how many people it can reach, how many will respond, how many will convert at the rates the business already sees, and what each conversion is worth in Contribution MarginRevenue left after every cost that varies with each sale.. Produce three versions, low, middle and high, by varying the assumptions that are most uncertain, and write each assumption down beside its source. The scenario planning framework is the discipline here: a range shows the decision-maker what is plausible, while a single forecast invites them to discount it. Where the investment adds to an existing channel, remember Diminishing ReturnsEach extra unit of spend or effort buys less extra result than the one before it.: the marginal returns and forecasting guide explains why the next unit of spend usually buys less than the average so far. The marketing budget calculator is a quick check that the spend and the outcome belong to the same plan.
Say how the result will be proved
Before approval, agree how anyone will know whether the investment worked. The strongest answer measures IncrementalityConversions caused by marketing.: what happened because of the investment that would not otherwise have happened. A holdout, a regional comparison or a staged rolloutReleasing a change to a growing share of visitors, pages or regions in steps and reading each before the next; a test only if groups are random and concurrent. can show that; the budget increase incrementality test raises spend in selected regions and compares them with matched regions left alone. Where that is impossible, name the best available evidence and agree it in advance. State the result in finance's terms: the ROI versus ROAS comparison explains why return on investment, built on contribution and all costs, is the measure for judging an investment, while return on ad spend serves day-to-day optimisation within a channel.
Work out payback and break-even
From the costs and the three outcome ranges, calculate how long the investment takes to pay back, and how much it must produce just to cover its cost. The Payback PeriodMonths of customer margin needed to recover the cost of acquiring that customer. is the time until the contribution it generates equals what was spent; break-even is the volume of sales or enquiries at which that happens. The break-even calculator gives the number of sales needed to cover a given spend at a given order value and margin. Show ROIProfit relative to cost. for each range, and say clearly whether the low case loses money. A case where even the low range pays back is easy to approve; one where only the high range pays back may still be worth doing, but it should be presented as a bet with a stage that limits the loss.
List risks and dependencies
Name what could stop the investment working or make it cost more: production running late, a platform rejecting the ads, the team not having time to run it, a supplier, a dependency on the website or the sales team, results arriving more slowly than the forecast assumes. For each, say how likely it seems, what it would do to the outcome and what you would do about it. Name the dependencies on other teams explicitly, with the person who will secure each. Finance reads an honest risk list as a sign the case has been thought through; a case with no risks reads as one that has not.
Stage the commitment with decision points
Break the investment into stages and ask for the first one. Each stage ends with a decision point where the investment continues, changes, pauses or stops, judged against criteria written now. The phase-gate reviews framework sets out how to write those criteria: specific measures with thresholds, the data source, guardrails as well as goals, and what happens if they are not met. Fit each stage's criteria to what it can show; a first quarter can show response and early cost, not steady-state profit. Staging turns a large one-way decision into a series of smaller ones and makes approval easier, because the decision-maker risks only the first stage.
Write the one-page summary
Put the case on one page for the decision-maker, with the detail attached. The page holds the decision sentence, the alternatives considered, total cost by stage, the low, middle and high outcomes with payback for each, how the result will be proved, the main risks, the first stage's decision point and its criteria, and what is being asked for today. Lead with the decision and the ask, not the background. Read it as a sceptical finance partner would: every number should trace to an assumption, and every assumption to a source.
Track the case after approval
Once approved, the case becomes the yardstick. Record actual costs and results against each range month by month, from the same sources the case named, and report them at each decision point. Where an assumption proves wrong, say so and update the forecast rather than waiting for the final review. Keep the case and its tracking with the other planning documents in the planning section, so the next case can start from what this one learned. A team that reports honestly against its cases, including the ones that did not work, finds the next one easier to get approved.
A landscaper asks for a video budget (illustrative)
A landscaping firm's case for project videos
Hedgeplume Landscaping, an invented garden design and landscaping firm, wins most of its work through search ads and referrals. Its marketing manager believes short videos of finished projects, run as paid video ads to homeowners in its area, would bring in larger garden projects. She writes a case for one quarter.
Decision. Approve £9,780 for a first quarter: film and edit six project videos and run them as paid video ads, with a decision at the end of the quarter. Alternatives: the same money in search ads, which are already showing diminishing returns at the current budget, or doing nothing.
Costs. Filming 6 projects at £600 each, £3,600; editing, £1,500; media, £3,000 over 12 weeks; her own time, about 2 days a month for 3 months at the firm's internal day rate of £280, £1,680. Total £9,780.
Upside, from the firm's own data. Last year 4% of website enquiries became projects, and the average project contributed £4,200 after materials and labour.
Assumptions and their sources. The visit-to-enquiry rate is 2.0%, 2.5% and 3.0% in the low, middle and high ranges (18 of 900, 45 of 1,800 and 81 of 2,700 visits). The 2.5% middle rate is an assumption made up for this example, and the low and high cases sit half a point either side of it. The visit counts come from the £3,000 of media at an assumed cost per site visit of about £3.33, £1.67 and £1.11 (£3,000 divided by 900, 1,800 and 2,700). In a real case each of these would be a figure from the firm's analytics or a platform estimate, written beside the number.
| Range | Video viewers who visit the site | Enquiries | Expected projects at 4% | Contribution | ROI on £9,780 |
|---|---|---|---|---|---|
| Low | 900 | 18 | 0.7 | £3,000 | −69% |
| Middle | 1,800 | 45 | 1.8 | £7,600 | −22% |
| High | 2,700 | 81 | 3.2 | £13,600 | +39% |
Contribution is worked from the unrounded project counts (0.72, 1.8 and 3.24) and rounded to the nearest £100. The break-even cost per enquiry follows from the firm's own data: one enquiry is worth 4% × £4,200 = £168 of contribution, so any cost per enquiry above £168 loses money. In the first quarter the all-in cost per enquiry, production included, is about £543, £217 and £121 across the three ranges (£9,780 divided by 18, 45 and 81), which is why only the high range comes out ahead.
In the first quarter even the middle range does not pay back, because production is a one-off cost carried by a single quarter of media. She shows a second quarter reusing the same videos, with £3,000 of media and £1,680 of her time but no production: on the middle range it adds about £7,600 of contribution, so the two quarters together return £15,200 on £14,460 of cost, paying back towards the end of the second quarter.
Proof. Ads run in two of the firm's four postcode areas; enquiries from the other two are the comparison. To keep the comparison fair she pairs the four areas by size and by last year's enquiries per thousand households, then draws one area from each pair at random to carry the video ads, so the strongest areas do not all end up on one side.
Decision point at week 12. Continue to a second quarter if enquiries per thousand households in the video areas are at least 30% above the comparison areas and the running cost per enquiry is under £120; stop if the lift is below 10%; otherwise run a further 6 weeks with new edits. Running cost counts the quarter's media and her time (£4,680) but leaves out the production already spent, because a second quarter would reuse the videos. The £120 line sits about 30% below the £168 break-even; the middle range clears it (£4,680 over 45 enquiries is £104) and the low range does not (£260). The owner approves the first quarter.
Common mistakes
- A request that cannot be decided. State what is being approved, for how long and with what first commitment.
- Counting media and forgetting people. Internal time, production, tools and fees are real costs; finance will add them if the case does not.
- One confident forecast. A single number invites doubt; low, middle and high ranges built from the business's own data invite a decision.
- Borrowed benchmarks. Industry averages describe other businesses; use your own conversion and margin figures.
- Revenue instead of profit. Judge the investment on contribution and all costs, in finance's terms.
- No plan to prove the result. Agree the measure, and ideally a comparison group, before approval.
- Asking for everything at once. Stage the commitment and ask for the first stage.
- Forgetting the case once approved. Track actuals against the ranges and report honestly at each decision point.
