Why we’re such bad estimators, and how to fix it

One of the enduring challenges we all face as project managers is estimation. Despite heaps of experience, most of us just don’t estimate well. Let’s look at the issues that make us such poor estimators and outline steps to enhance our estimation capabilities.

Issue 1:

We tend to be too optimistic. In our enthusiasm to launch projects, and please our managers, we get too optimistic about the time to complete tasks and underestimate the potential roadblocks. We also think about the total heads-down time it may take to complete tasks. While doing so, we disregard the interruptions, competing priorities, and other items that take team members away from our projects. This produces overly optimistic project timelines.

The fix:

• Remember history. Never assume a team member will work 100% of their time on your project. Think about the interruptions that intruded onto your prior projects.
• Leverage PERT estimates. Asking your team members for the optimistic, most likely, and pessimistic estimate for how long something will take is fruitful. It generates more thought and surfaces possibilities that might otherwise be forgotten if you ask for a single estimation number.
• Create a more granular WBS. The smaller the task to be sized, the easier it is to estimate. A more granular WBS produces more focused tasks, making it easier to produce an accurate estimate.

Issue 2:

A lack of historical data. Note in the last item I said to “remember history,” not look up your history. This is because very few organizations capture task durations and effort levels to help with future estimating. While projects are unique, there are often similar tasks that are repeated when executing projects that can be used to improve our estimation.

The fix:

• Capture your own historical database. Don’t wait for your organization to wade through “how to” approaches and design a database to cover the entire business. Start compiling your own historical data. Capture task durations and the effort required to accomplish common tasks, like creating a business process, or scoping out a new IT tool.
• Apply and modify. Reference the history you compiled and compare the results to your current project’s tasks. Add or subtract a factor based on complexity and/or the level of skill you will have to complete a task. This will help you create more accurate estimates.

Issue 3:

We allow too much scope creep. Be fanatical about managing your scope. It’s okay if the business comes up with new or changed requirements. But, be ultra-diligent about examining and estimating the tasks that will be added to your project.

The fix:

• Implement a strict change control process. Evaluate every proposed change for cost, complexity, staffing, and risk. Ensure a detailed evaluation, with a minimal amount of conjecture, is brought forward for approval or rejection by the sponsor and/or product owner.
• Determine a ceiling for accumulated changes. Often, scope creep still occurs, despite a diligent change control process. Each change might stand on its own merit. However, the cumulative effect of approved changes could significantly change the direction and purpose of the project. Should your project changes add 20% to your project timeline or budget from its original incarnation, this should trigger additional reviews. A project that changes that much might require a thorough evaluation, and potentially be re-conceived, or broken into a series of smaller projects to minimize risk and realign estimates for accuracy.

Issue 4

We rely on uneducated estimators. On many occasions, team leaders provide estimates, but a different team member will perform the project task. When the time to complete the task doesn’t match the estimate, there is rarely a conversation between the team leader and team member to discuss the situation.

The fix:

• Educate estimators! As a matter of course, whenever tasks are completed, project managers should facilitate a discussion about the estimate, and what it took to complete the task. Without this feedback, the estimator won’t adjust their estimates in the future. Through a lack of education, they consistently provide inaccurate estimates.
• Listen and adjust – scientifically! The benefit of having project managers facilitate the discussions to educate estimators is that they become more educated as well. And, when an estimate is provided that doesn’t match that learning, the PM can make an educated adjustment. Doing this is much better than applying a blanket and totally unscientific adjustment of doubling the estimates received. While this might work in many instances, it isn’t defensible, doesn’t promote learning within the organization, and can cause a valid project to be rejected as being too expensive when that isn’t the case.

Issue 5

We take a one-and-done approach to estimating. Many project managers provide an estimate for their projects, and then proceed without updating their estimates. Despite the Project Management Institute’s (PMI) guidance to provide three levels of estimate, an initial estimate is all that is provided. That estimate is rarely accurate but isn’t revised when more facts are known.

The fix:

• Utilize Order of Magnitude, Budgetary, and Definitive estimates. Using these progressively more accurate estimates provides more accuracy, based on facts that surface during the project. For example, the cost for a vendor contract might be an estimate during earlier stages of the project. However, when the vendor contract is negotiated and signed, a definitive (or more definitive) cost can be applied to an overall project cost estimate.
• Label your estimates. Don’t just provide a number, such as “The cost estimate is $560,000.” Rather, share the stage of your estimate and the variance associated with that estimate. For example, “The BUDGETARY cost estimate is $560,000, with a project variance of +25% to -10%.”

_____________________________________________________________

Additional thoughts can be found in my project management and outsourcing classes on LinkedIn Learning, including:

• Project Management: Technical Projects
• Strategic Leadership: Deploying Intelligent Disobedience
• Outsourcing Fundamentals , the first of six courses on outsourcing. This series of courses focuses on defining services, performing service-level?management, and helping figure out what is required to support a business in an outsourcing or service provider scenario.

_____________________________________________________________

Coming Up

Office Hours Live - Creating High-Performance Teams

Leaders aren’t really leading unless they have a team who wants to follow them. When the team works effectively, project performance is amazing. Creating high-performing teams - particularly when you rarely get the chance to choose your team members – is a challenge, and also a real opportunity to work your leadership magic.

Join me and Mark Vollmer, an expert on building peak-performance teams, as we explore ways to help you create high-performing teams . Come along and bring your questions – we’ll answer them LIVE.

By attending this session, you will:

? Understand what makes a team high-performing.

? Reinforce what you know and build your team leadership capabilities

? Learn tips for building an environment where a high-performing team can emerge.

? Explore ways to assess your team on the journey towards high performance.

? Assess your own leadership style and how it can help drive or limit peak performance (or not!).

? Discuss three things you can do TODAY to help your team improve its performance.

_____________________________________________________________

This article is part of Bob’s Reflections newsletter series , which discusses project management, outsourcing, and “intelligent disobedience”, a leadership approach. If you want more of this content, you can subscribe to receive notifications when a new article is posted.

Want to learn more about the topics I talk about in these newsletters? Watch my courses in the LinkedIn Learning Library or check out https://intelligentdisobedience.com/

_____________________________________________________________