Forecasting under Uncertainty

By Don Creswell, Co-Founder SmartOrg Inc.

Let’s start with an emphatic statement: there are no facts about the future, therefore you cannot forecast with accuracy. The longer the planning horizon the less precise your forecast will be. Hit your numbers and you are lucky. No wonder so many projects and products fail to meet projections or fail altogether—as high as 70% say some surveys. That’s a lot of money and, most likely careers, “down the drain.”

“But, you say, I have to forecast sales, revenues, profits and so on. If what you say is so, how do I deal with this?” Good question. In the next issues of ValuePoint, I will provide answers to this question, based on several decades’ experience where uncertainty abounds: R&D and new product development.

People aren’t dumb. One way to get around uncertainty is to make conservative forecasts, raising the probability that you will win, but killing innovative growth. Another way is to inflate your numbers, get the project going and hope for the best. Good for the individual; bad for the company. Yet another way is to develop highly sophisticated economic models and simulations that provide a degree of comfort and a shield when the numbers fail to pan out. Unfortunately decision makers have no clue as to what is in the models and either take results as presented or disregard the numbers and go with their gut feel.

SO WHAT DO YOU DO?

First, acknowledge that you cannot put a single number on a future event. If you are the boss and someone comes to you with a sales projection for several years out that consists of a single number—especially if that number goes to the first or second decimal place—toss him or her out of your office. Same deal if someone tries to snow you with complex spreadsheets that you do not understand.

Second, recognize that all numbers about the future are uncertain to a certain degree; some are very uncertain like the size the market and the share you can obtain for a new product. Others are less uncertain, like the cost of a product where you have decades of experience.

Third, understand that the structure of business models used to develop forecasts at the planning level can be—and should be—relatively simple. Let’s look at a typical model for a new product:

There are three components that make up the model:

Technical: “Can you do it?”
What is the probability that you can successfully complete each phase and proceed to the next phase? What is the range of cost and time it will take (duration)?

Commercial: “Given that you can do it — what is it worth?”
Answering this question involves identifying factors that contribute to developing cash flow; each will require assessing a range of values depending upon your current knowledge, experience and available data.

Launch Costs and Time

The Complete Model

The blue rectangle represents the cash flow developed from revenues and costs. The yellow ovals represent calculations from the technical, commercial and cost sections of the model. The red octagon is the result of hundreds of calculations that develop the expected net present value; NPV weighted for the impact of uncertainty.

While this appears to be a fairly simple model, the model basically contains all the factors you need to develop a forecast of the economic value of a project or product.

How does the model handle uncertainty?
Where do the numbers come from? What does the computer do? What information and direction does the computer provide from the evaluation? What is the math underneath?

These questions will be addressed in the April 2010 issue.

Screen shots in this article are from Portfolio Navigator™, web-based software that supports evaluation of the economic value of projects, products and project/product portfolios.

To learn more about Portfolio Navigator, please contact info@smartorg.com or visit www.smartorg.com.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Free Briefing and Software Demonstration
If your company is challenged to improve the return on investment in projects and product development, let us show you how you can raise your chances for success through value-based management process and software. To schedule a free, private web briefing contact jholmes@smartorg.com.