The one-page business case a CFO can actually sign off.

Finance leaders often receive two versions of the same request.

🙁 The first is a polished deck forecasting uninterrupted growth and asking for another £50K of media spend. It leads with a confident headline number but gives finance very little on which to base a decision.

🙂 The second sets out the investment, expected return, payback period, confidence level and the point at which spending will stop. It is honest about which numbers are proven and which are still estimates. It gets backed because it gives finance something it can responsibly stand behind. It answers the questions the business will ask when the results come in.

Why the polished case comes up short

Marketing cases are often built around reach, acquisition and platform-reported return. Finance has to assess the same investment through cash, contribution and downside exposure.

The case arrives built around one set of numbers and gets read against another. Finance cannot find what it needs, asks for more detail, and those questions can come across as finance getting in the way.

Round it goes, with the finance leader labelled as the person who does not understand growth.

The polished case is not necessarily dishonest. It is incomplete.

It shows the opportunity but leaves out the two things finance most needs to know: when the money comes back, and what happens if it does not.

Without those answers, someone still has to decide where the spending stops. By default, that responsibility falls to finance.

The solution is a shared understanding of what a complete business case looks like. In most cases, everything finance needs to make the decision can fit on a single page.

What belongs on the page

How much money, and over what period?

State a clear investment amount and spending period.

Not a range that quietly doubles once it has been approved.

When does the money come back?

Set out the expected payback window.

A pound spent in January that returns in March is a very different proposition from one that returns in October. That difference is not a detail. It is cash flow.

What is the real return?

Do not rely solely on the gross return reported by the advertising platform.

Show what the business keeps once the product, fulfilment, shipping, fees and media costs have been paid.

This is the gap covered in the first article in this series: a reported 19x platform ROAS that translated into a 4x contribution return. Read more on this here.

The case should also show the downside. What happens if acquisition costs are 20% higher, conversion is weaker or repeat purchase takes longer than expected?

A simple base case, upside case and downside case is often more credible than one precise forecast.

How confident is the team in the numbers?

This is where many cases stumble, usually by overclaiming.

A proposal that says, “High confidence, because we have run this five times before,” is stronger than one that claims certainty over a result nobody can guarantee.

Perfect attribution is rarely available, and waiting for it can become an excuse not to learn.

Good attribution, supported by an honest confidence level, earns finance sign-off faster than a flawless model that took three months to build.

What evidence sits behind it?

A relevant comparable can do much of the heavy lifting.

That might be:

• We did this last year and this is what happened.

• The last five times we ran it, it worked within this range.

• This customer cohort behaved in a similar way.

• Another comparable business tested the same hypothesis and saw this result.

External examples do not prove that the same outcome will happen in your business. They can, however, show that the hypothesis is reasonable enough to test.

Evidence is what turns a request into something finance can responsibly back.

What are the signals to scale, pause or stop?

This is the line that separates a request from a proposal.

Before a penny is spent, a strong case should define what good looks like, what bad looks like and what the team will do in each scenario.

A case that names its own kill signal has already answered the question finance will otherwise have to ask.