Guide

How to justify a capital investment to the board

A board approves capital when it can see four things: what the money buys and when, what the alternative to spending it is worth, what the case looks like if the figures come out badly, and where every assumption came from. Bring a range rather than a single number, price the option of doing nothing or waiting, name the two or three figures the answer depends on before anyone asks, and state the downside in money on the same page as the upside. A case that brings its own weak points is approved faster than one that hides them.
Run the live demo

What a board is actually doing when it reads your case

It is looking for the part that does not hold. Not out of hostility: a board that approves capital it does not understand has failed at its job, and the members know it. So the case is read the way an auditor reads a reconciliation, from the weakest number outward.

That tells you how to write it. A case built to persuade invites the search for the hidden flaw. A case built to be interrogated, that hands over its own weak points with the figures attached, ends the search early.

Step 1. Price the alternative, including doing nothing

Every capital case competes with not spending the money. Put that option in with its own figures: what the business does with the capital instead, what deferring two years costs in benefit forgone, and what the capital is likely to cost when it is finally spent. Deferral is the option boards reach for when a case feels thin, so arrive with it already costed.

Step 2. Bring a range, not a point estimate

The single number in a capital case is always the one that clears the hurdle, and everyone in the room knows it. A range is both more honest and more persuasive, because it shows the work: here is the low case, here is the middle, here is the high, and here is how often this clears the bar.

  • Capital cost, as a range that includes the figure the engineers actually gave you.
  • Benefit a year once it runs, as a range.
  • Months to build and commission, and slippage on top of that as its own range.
  • The period judged over, and the discount rate, stated once and used for every option.

Step 3. Name the figures the answer depends on, first

Work out which two or three inputs move the comparison, and put them on the page before the board finds them. This is the single highest value thing in a capital paper: it directs the discussion at what matters and it tells the board you know where your own case is thin. If slippage is on that list, say so, and bring what you are doing about it.

Step 4. Put the downside in money, on the same page

A risk section written in adjectives is not a risk section. Convert it: if the build slips six months, the case is worth this instead of that. If the saving comes in at the bottom of the range, here is the number. Boards do not need to be protected from the bad case; they need to know it is bounded and that somebody has looked at it.

Step 5. Make every assumption traceable

Each figure should be attributable: who gave it, from what, on what date. An assumption with no owner is the first thing a board distrusts, correctly. This is also what makes the review in two years possible, which matters more than it sounds: a company that can tell a bad decision from bad luck gets better at both.

Step 6. Be able to answer what if in the meeting

The question will come, and the gap between answering it in the room and answering it next week is the difference between an approval and a deferral. If changing an assumption means a week of rework, you will present the version that cannot be questioned, and that is the version boards approve least.

Doing it with BrainPower AI

The product covers steps two, three, four and the in meeting part of six: ranges instead of point estimates, a low case, a middle outcome and a high case per option, the chance each one comes out ahead, the figures the answer depends on ranked, and a rerun in seconds when someone changes an assumption. See the project approval use case for a worked example, and capital allocation for the case where several programs compete for one budget. Steps one and five remain yours.

Questions people ask

What makes a board reject a capital case?

Most often not the return. It is a case that cannot be interrogated: a single number with no range, no priced alternative, a risk page of adjectives, and assumptions nobody owns. A modest return that is clearly understood is approved more often than a strong one that cannot be checked.

Should I show the internal rate of return or the net present value?

Show what your board uses, and show the money. A return percentage without the amounts hides the size of the bet, and an amount without a period hides the cost of waiting for it. If you discount, say what rate you used and use the same one for every option.

How do I handle an assumption I cannot support?

State the range you can support and show whether the decision depends on it. If it does, the honest recommendation is to go and find out before approving, and boards respect that more than a confident figure that turns out to be invented.

How much detail should the pack carry?

Two pages of substance: the options with their low, middle and high cases, the figures the answer depends on, the downside in money, and who owns each assumption. The detail belongs in an appendix nobody reads unless they want to.

Does this work for an investment committee as well as a board?

Yes, and committees tend to press harder on the alternative use of the capital, so step one matters more. Price the alternative properly and the rest of the conversation goes faster.

Keep reading

See it on your own numbers

The live demo runs in the browser with figures you set. A demo with us walks through justifying capital to a board on a decision you are actually facing.

Run the live demo