Use an appropriate discount rate.
The discount rate used should reflect the opportunity cost of the capital required for the investment. Now, this is easier said than done and depends on a few factors like the applicable tax regime or the risk associated with the discount rate. In any case, the discount rate should be higher than just the cost of a corporate loan to finance the project. Ask your CFO, or contact a specialist!
Consider cashflows, and only cashflows
The NPV looks at cash, not at any other accounting metric like revenue or profit or capex. So make sure to identify all cash flows relative to your investment and make sure to be precise on timing of the cash flows. If a capex budget is spent over the course of two years, it is an oversimplification to just account for one cash out at the start of the project.
Discount to the moment of decision making
The NPV of an opportunity corresponds to the value created (or destroyed) if you decide to do it. So the value is created at the time of decision. Therefore you have to discount to the time of decision, often “today”. Hence, discounting to the moment of the first actual cash out related to the investment is therefore an overestimation of the value of your project.
Only consider incremental cashflows
You should only consider the cash flows that change the future cash position of the company as a consequence of the investment decision. This is maybe one of the most difficult tasks in the whole process. You have to compare the future company cash flows in case you decide not to invest (which is already a tricky undertaking) with the future company cash flows in case you do invest.
This also means that you should not consider unavoidable costs and sunk costs. If your decision has no impact on the cash flows, they should not appear in your model. Easier said than done, again.
Consider all expected cashflows
Another tricky element is that you have to identify all of the cashflows related to the decision. That means for example also cash flows in the fiscal domain, something engineers sometimes tend to forget.
A special case is the residual value of the asset you are investing in. It should reflect the value of the asset at the end of the considered time horizon. Setting it to zero is often an underestimation.
Real or nominal value, as you please, but be consistent
Choose for real values (without inflation) or nominal values (including inflation), but don’t mix them up and make sure your discount rate is in accordance.
Don’t forget the real option value
What are your “real options” related to the investment. Can you stop it prematurely? Can you defer it in time? Will it create another opportunity which might not exist without the initial investment? A decision which does not take into account the value of the options will almost always underestimate the value of your business case.
And many more
So easy the math is, so difficult it can be to define a consistent NPV which will lead to optimal decision making. Which risk is not accounted for in the discount rate? How to handle that? How to compare or prioritize different alternatives if they have a different risk profile of lifetime?
If you want some sound advice, give us a call and we will see how we can help you drive your company’s decisions.