From the course: Project Management Foundations

How to choose the best estimate

From the course: Project Management Foundations

How to choose the best estimate

“

- Project estimates are a lot like your morning commute. You need to choose an estimate that gives you a good chance of getting where you need to be when you need to be there. On some days, your commute is a breeze, maybe 30 minutes. Most days, let's say it's 45 minutes but an accident can turn it into an hour. For your commute, you choose a drive time that gives you some wiggle room so you get to work on time. The same approach works equally well for your project estimates. You probably decide when to leave for work informally. We're going to look at a similar, yet rigorous approach to choosing estimate values for your project. If you were to graph the probability of your commute times it would look like a bell curve with your average drive time at the peak. A bell curve is also known as a normal distribution and it has properties that help you choose a safe estimate to give your customer First on a bell curve, 50% of the possible values are less than the average, and 50% are greater. That means the average value gives you a 50 50 chance of your results being equal to or less than your estimate. Think about it. If you give your customer your average estimate you are as likely to fail as you are to succeed. I wouldn't stake my career on those odds. On the other hand, the worst case estimate gives you the highest chance of success. However, that number is so high, your customer might cancel the project. You probably know that customers are prone to ask what's the best case scenario? What if everything goes right? Don't fall for this trap. Suppose you give the customer your best case estimate say six months, and you describe everything that has to go right to finish the project in six months. This is what they'll hear, blah, blah, blah, blah six months, and you've set up your project to fail.. We've eliminated the best, worst, and most likely values. You're probably wondering what number you should use for your estimate. The answer is about halfway between the average and worst case values. Let's go back to the bell curve and pick the value halfway between the average and worst case values. Without getting into the mathematical details, that gives you an 86% probability of your results being less than or equal to that number. That's a good chance of success, and the value is easy to calculate. Plus, you can tweak your value higher to reduce the chance of missing the deadline, just like you would leave extra early if you're going to a job interview and if you're trying to win a project you might choose a smaller value even though the risk of failure is higher. The bottom line is choose the estimate that provides an acceptable probability of success. For practice, use the information in the exercise files to determine the time estimate you would give the hospital COO.

Contents