A Better Way to Avoid Project Delays

Why do so many managers underestimate the time it will take to complete a significant project?

Almost half of business projects fall behind schedule, and up to a third are not completed at all, according to a 2021 report by the Project Management Institute. The Berlin Brandenburg Airport took 14 years to construct instead of the estimated five, and its cost ballooned from a 2009 budget of 2.83 billion euros to over 6 billion euros by the time it was completed. And after the V.C. Summer nuclear plant expansion project in South Carolina was abandoned in 2017, two executives involved in what became known as the “Nukegate” scandal pleaded guilty to fraud charges stemming from their efforts to conceal delays.

Late projects trigger cost overruns, reduce quality, and upset clients. While some projects fail due to poor execution or changes in scope, a frequent culprit is an unrealistic time frame that dooms the project before it even starts. Nonetheless, planners often set project schedules that are too short — partly because of uncertainty and ambiguity around the full scope of work required but also due to the expectations and enthusiasm of project sponsors.

Excitement about a new project can spur optimism, which may bias even experienced executives to underestimate the time needed to execute the work. However, if members of a project team consider a CEO’s optimistic estimated time to complete to be plausible, that is extremely likely to influence their judgment and be reflected in their own estimates. This is an example of anchoring, a cognitive bias that causes us to give more weight to the first piece of information we get on a given topic.

Unrealistic schedules may be introduced by a variety of unintentional anchors, including initial wild guesses (“Maybe we could get it done in two months?”), suggestions (“Can you complete the task in three weeks?”), customer expectations (“We’re planning to launch the product right after the new year”), and tentative deadlines (“The CEO asked for the project to be completed by the end of next month”).

Our own research confirms that anchors influence schedule estimates and that their effect persists over time. In a controlled experiment, we asked 93 participants to complete one task, repeated three times, after giving an advance estimate of how much time it would take. Two groups of participants were exposed to anchors before making their estimates: We asked them, “Will it take you less or more than [our anchor value] to complete the task?” Participants in one group were shown an anchor value of three minutes while those in the other group saw a value of 20 minutes. A third group, the control, was not exposed to any anchors before making an estimate.

Compared with the control group, on average, people who were exposed to the high anchor repeatedly estimated that it would take them longer to complete the task, whereas the low anchor led to shorter time estimates. The participants were not told how long they took to complete the task, and the bias persisted on multiple rounds of estimating and executing the task. Importantly, the actual average task duration did not differ between groups.

We also found evidence of self-anchoring within the control group: Control group participants who underestimated task duration in the first round kept underestimating it in the following rounds, and those who overestimated it also continued to do so, even though the time it actually took the two subgroups to execute the task was the same.

Thus, to elicit unbiased project schedules, it is crucial to isolate planners from uninformed guesses or suggestions.

Using Anchors to Improve Estimation

Given the power of anchors to influence estimates, can project managers strategically use them to avoid overly optimistic schedules and induce people to create more accurate timelines? Project managers often hope to achieve schedule accuracy by collecting excessively detailed descriptions of project deliverables. While more information does result in more realistic schedules, the gains are often marginal and project plans remain too optimistic. Project planners intuitively focus on available specifications, usually not realizing that even the most comprehensive descriptions are necessarily incomplete. In addition, excessive details may lead to more confusion.

Our second study, which involved 139 new participants, compared the effect of a detailed project description with that of a “helpful anchor” (the average duration of similar, past projects, also known as reference class or outside view information) on timeline estimates. We found that both interventions mitigated underestimation: The average actual performance of both groups more closely aligned with their average estimated performance times than was the case for a control group that had received no assistance. However, the estimates produced under the detailed description had a large variance. While the group average was relatively accurate, individual participants tended to provide large over- or underestimations. In contrast, estimates produced with the helpful anchor were accurate not only on average but often also at the individual level.

Our experiment showed that anchoring planners on reference-class information can improve schedule accuracy and on-time project completion. This is a simpler and more effective approach than giving planners as much information as possible about a project.

On Time Doesn’t Always Mean Successful

When project managers are free to set their own timelines, judging their success solely based on their ability to execute projects on time and within budget can have unintended negative impacts on company performance. In the planning stage, project managers may inflate their time estimates and budgets to levels they believe are still acceptable to stakeholders. Later, they may deliberately slow the project’s execution to match the agreed-upon deadline. Although the finished project is delivered on time, the inefficiency of having been delivered later than feasible remains hidden. On top of the wasted time and resources, hidden inefficiency frequently prevents organizations from engaging in new projects and results in lost revenue.

Our most recent research found that incentives also play an important role in project estimation and delivery. We tested whether focusing exclusively on timely project delivery induces inflated estimates and slower project execution. In the experiment, 198 new participants performed a simple search task. To establish a benchmark for task completion time, the first group was incentivized solely to complete the task as fast as possible. The second group, which was incentivized only to make the most accurate estimates, provided significantly higher projections and then slowed its work pace to finish close to its estimates, as we had hypothesized the group would. Participants in the third group were rewarded for both their speed and accuracy; they completed the task much more quickly than those in the second group and often made accurate estimates. These findings demonstrate that project owners must design incentives carefully because they can backfire and slow down project execution.

Based on our research, we recommend that project owners take these steps to improve the accuracy of schedule estimates:

• Take care to eliminate poor-quality external anchors, such as the CEO’s hopes and expectations, during the estimation process.
• Carefully consider past experiences, and anchor estimates on historical project data to complement detailed project plans. A measure as simple as the average duration of similar completed projects often outperforms the accuracy of planners’ estimates.
• Avoid placing a strong emphasis on avoiding schedule overruns. Focusing exclusively on timely delivery can backfire because people may strategically inflate estimates and slow their work pace to match them. To mitigate the hidden inefficiency, organizations must align the objectives of project owners with those of planners. For example, emphasizing or rewarding performance speed on top of estimation accuracy can lead to compressed but reliable schedules and faster project completion, without compromising the quality of the deliverables.

---

Delays, missed deadlines, and overshot budgets are some of the most common problems projects face. If project planners and managers are more aware of the strong effect that anchors have when estimating timelines and budgets, they can avoid being led astray by uninformed anchors and use helpful anchors to set more realistic expectations.