William W. Davis, MSPM and PMP, Project Management Superhero Project managers project the future every day. We plan for the future, we schedule tasks, budget for Quarter 4, and we book resources to support activities that haven’t yet happened.
But how much do we really know about the business and science of forecasting the future? For a quick primer on forward planning, I spoke with William M. Davis, a forecasting genius.
William, what is the difference between forecasting and predicting the future?
Weather forecasting is a concept you might consider. Weather forecasts and predictions can be made by your favorite meteorologist or smartphone weather app.
A prediction is a single-value, deterministic estimation of future uncertainty. A forecast is a probabilistic estimate of the possible outcomes of future uncertainty.
I don’t get it!
Weather forecasters predict high temperatures and low temperatures for the next day. For example, South Florida’s daytime high temperature prediction for tomorrow is 90 degrees Fahrenheit.
Although I know it might not be exactly 90 degrees tomorrow I have an intuitive tolerance of error for predicted temperatures. Tomorrow’s temperature could rise to 88 degrees or rise to 92 degrees. It’s going to get hot, regardless of what.
Meteorologists can forecast the likelihood of rain tomorrow.
Tomorrow’s rainfall forecast is for 50% of the area where I live. Rain tomorrow is possible, but not probable.
It might rain tomorrow, but that doesn’t tell me much. Weather forecasters use a percentage to express the likelihood of rain.
It’s done! Which one is better?
Both forecasts and predictions are useful!
Predictions are easier to make and share when we need quick estimates of the future. However, predictions that are based on uncertainty can be dangerous if there isn’t a common understanding of the risks involved.
We should not predict, but forecast in order to convey our uncertainty.
This is fine for the weather, but what does it have to do with work? How can forecasting help me better manage stakeholder expectations
Have you ever given a project forecast for cost or schedule early on in a project’s lifespan, but the sponsor took those early predictions as firm commitments.
Hasn’t everyone? !
It’s better to create a forecast if you need to convey uncertainty about your estimate. A bell-shaped probability curve is often used to communicate the probabilities of an uncertain future.
Forecasting allows you to make statements such as these:
“With 80% confidence, that project can be completed for $300,000.
“If we want 95% confidence, we will need a budget of $340,000.”
“Yes, we can reduce the budget by $30,000. “Yes, we can cut the budget by $30,000. But I’m only 60% certain that the project will not exceed its budget. Are you willing to take that risk?
How about helping my sponsor/organization make better decisions?
Let’s say you are my project sponsor and you want to get an early, high-level estimate of the cost of the project we’re about start. Let’s say you are willing to invest as much as $1M in this project.
Would you consider funding the project if I gave you a $900,000 project cost prediction, which is my “most likely” outcome?
Most likely. My cost prediction is lower than your spending threshold. However, my cost prediction does not convey any uncertainty I have about the project’s “most likely” outcome.
Let’s say I created a cost forecast for a project. This might look something like this:
It is easy to see the cost uncertainty associated with your project. Even though $900,000. is the most likely outcome, there is still a chance that your project will exceed your $1M funding threshold.
But what risk is there? Let’s take a look at the same forecast but in a different way:
Yikes! My forecast shows that there is a 29% chance (the orange pie slice) that the project might cost more than you anticipate.