Accurate Project Estimating – Video

Dick Billows, PMP
Dick Billows, PMP

Making accurate estimates of project duration, cost and completion date are critical to your project manager credibility and career. No matter how good your technical skills and judgment are, if decision-makers can’t count on your estimates of cost and duration, your credibility will inevitably suffer. Accurate estimating is not primarily a matter of knowledge or intelligence.  Making accurate estimates requires the correct techniques and the ability to decide when to use each project management estimating technique.

Here’s a video on estimating techniques.

The Right and Wrong Way to Estimate Task Duration

Estimating Data

Good data, an important piece of information we need for estimating, is often the most difficult is to get. We also need to know how to use the data once we get it. There is no need for sophisticated statistical techniques or databases. I’m talking about assembling estimating data from one or more readily available sources. The first source of this estimating data is historical information from previous projects. This is the easiest to get of the three sources of data and it costs very little. In fact, you can review the data from your previous projects and have a decent start. The data is better if all the project managers in the organization archive their project data. The estimating technique that uses that historical data is called analogous estimating.  Gathering the data is a simple a matter of using your project management scheduling software to collect it. You need the project software to store data about the actual hours of work and cost on each project.  You also need to use the software to create a schedule based on hours of work, not by plucking completion dates out of the sky. If you have this data for every project you manage, you’re off to a great start.  If the organization has every project manager generate this information on all projects, estimating becomes much easier and much more accurate. It’s simply a matter of archining the data on completed projects.

Another source of estimating data is published information, particularly in certain industries like construction and information systems projects. The estimating technique that uses this published data is called parametric estimating. There are some downsides to the published estimating data that we will discuss below; but don’t overlook it.

The third source of estimating data is the project team members who are doing the work on your project. The estimating technique that uses this data is called 3-point estimating. It is also called PERT estimating, which stands for Program Evaluation and Review Technique. You must properly train and manage your people in estimating the cost and duration of their work.  Their 3 estimates of the work – optimistic, pessimistic and best guess – will give you a valuable source of information. Getting estimating data from the project team members increases their commitment to those estimates. It also supports the team building that goes along with the team members participating in the estimating process. But you must also train yourself to accept their estimates and any errors. If you become an hysterical maniac every time somebody exceeds their estimate, all you will do is train the team to pad their estimates.  But if you treat each estimate overrun or under-run as a learning opportunity for the team member, you can end up with very good estimators who don’t play games with the process. Professional firms do this very well and that is why they can make money on almost all their projects.

Every project manager needs to have several estimating techniques in his or her project toolkit. Another technique is used during initiation when very little is known about the project. It’s called order of magnitude estimating and executives and project sponsors hate it. That’s because the order of magnitude estimating has ranges of numbers. It’s entirely appropriate at the very beginning of the project when we have little knowledge about the nature of the project and the level of risk.  The PM is in the dark about the required work and the resources available to do it. A project manager would be foolish to estimate the duration or budget that early in the project. If pushed by the executives, the PM should very carefully hedge their bets by saying, “This project may take between 100 and 200 days.” I think you can see why executives hate that kind of estimate.  When they hear this, they remember the low number and forget the high number. So you should always put estimates in writing.

What Estimating Techniques To Use 

Which estimating technique to use at each point in the project lifecycle depends on a number of factors. First, the scale of the project and its importance to the organization. That will affect how much time you spend on the estimates. On a large project, you might invest substantial time in unearthing historical data to use in analogous estimating. On a small project, you might make quick 3-point estimates with the project team and limit the effort to that. You can also use parametric estimating and rolling estimates weekly as the project progresses. We apply those at specific stages of the project lifecycle to give the project sponsors the data they need for decision-making.

Project managers should use these estimating techniques starting during initiation, then during high-level and detailed planning. They also use them weekly during the project to give executives accurate data about the project’s duration and cost. The key here is giving the executive the best data you have, while recognizing that your certainty about the cost and duration is the worst at the beginning and gets better as you progress through planning and then begin to implement. By reporting your confidence in the numbers, you can give estimates without making commitments that you can’t deliver.

During initiation, you would use analogous estimating based on historical data from other projects that are similar to the one currently working on. The estimates you would give at this point might be within the range of -25% to +75%. As an example, you would say that a duration estimate might be a range of 75 to 175 days. That confidence range would narrow as you move through the planning process. They would narrow further when you finalize the plan.  Each week as you execute the range the estimates would with become tighter and tighter until you are able to offer a -5% to +10% estimate of duration. In the example, that might be 95-110 days. As you know more about the project, these rolling estimates can become increasingly precise.

You can learn proven techniques for accurate estimating in our online project management courses. You’ll work privately and individually with a expert project manager. You control the schedule and pace and have as many phone calls and live video conferences as you wish.

Get free articles and videos like this every week

Author: Dick Billows, PMP

Dick has more than 25 years of project and program management experience throughout the US and overseas. Dick was a partner in the 4th largest professional firm and a VP in a Fortune 200 company. He trained and developed 100's of project managers using his methodology. Dick is the author of 14 books, over 300 articles and director/producer of 90 short project management training videos. He and a team of 25 project managers work with client companies & students across the US and in Europe, South America, Asia and the Middle East. They have assisted over 300 organizations in improving their project performance. Books by Dick Billows, PMP are on

3 thoughts on “Accurate Project Estimating – Video”

  1. Pingback: Work Package
  2. This is a very interesting approach but the way we do it here is to back into a completion date. What we do is the sponsor gives us the completion date and then we figure backwards and decide when the last task has to be done and then figure when the one before it has to be done. We keep going until we get to the first task. Most times we see that we should have started three weeks ago. How do I get the sponsor to change?

    1. Jake, The big issue here is if the projects are finishing later than the sponsor wants. If it’s working and you are somehow finishing by the sponsor’s date, he/she will never change. If the projects are late, however, you need to use the argument that improvement will come from better estimating rather than pretend due dates.
      Want does everyone think?

Comments are closed.