Devils in detail and project estimating fallacies

Project estimating costs and durations can be done in great detail using Excel or more sophisticated tools. However, detailed is not synonymous with accurate! Excessive detail can reduce accuracy, devalue the estimate and create unrealistic expectations leading to failure when the project fails to achieve the impossible.

Unless the work is designed in its entirety and all subcontractors and specialists appointed before any work commences, it is impossible to accurately estimate the work in its entirety, in detail at the beginning of a project.

The Practice Standard for Project Estimating defines estimating as ‘The act of creating a quantitative assessment of the likely amount or outcome’; by definition, you estimate future work and therefore the actual outcome is unknown and uncertain. And importantly, you have no idea how uncertain: past performance is only an indicator of potential future performance, there are no guarantees!

Where this paper disagrees with the PMI standard is in the assumption that ‘The most accurate and most reliable estimate for a project can be developed with the use of a bottom-up estimating technique. Prerequisite to a bottom-up estimate is a detailed WBS [work breakdown structure], and a comprehensive list of project resources.’

Bottom up estimating is only more accurate for work in the short term where all relevant information is available and current. This means the technique works for small internal projects with a maximum duration of a few weeks and for the near term work on larger projects where the people involved in doing the work, the actual complications and degree of difficulty of the work and the current performance levels are known.

Greater accuracy can usually be obtained on larger projects using other techniques such as vendor bid analysis, parametric or analogous approaches. Adding inaccurate detail simply confuses the estimate without improving the accuracy of the overall answer.

Pragmatic estimation

Early in my career I had to estimate the time needed to plant 35,000 plants on a rocky hillside as part of the construction work for the Argyle Diamond Mine accommodation village (shown below).

With a labour rate of $60 per hour, every minute spent planting a plant added $35,000 to the project cost. With an expected crew of 15, the task duration changed by five days. After spending several days working on a second-by-second estimate to determine the accurate planting time for each type of plant and shrub, we abandoned the exercise and focused on understanding how long similar jobs had taken and how the work crews were organised. An analogous approach produced a more accurate and stable estimate.

The generation of a details estimate also needs to be managed from a perception/expectation perspective. An estimate of $10,988,547.55 is no more valid than the estimate stated in more realistic terms such as; $11 million with a probable range between +10% and -5%. The precisely wrong number calculated to the nearest cent will raise the expectations of a range of stakeholders as to degree of accuracy that can be achieved in an estimate, leading to ‘perceived failure’ when the stakeholder’s unrealistic expectations are not realised. It only takes a cost increase of $2,000 (an estimating error of 0.02%) for the project to ‘fail’ because the costs have ‘blown out’ to more than $11 million.

The concept of ‘schedule density’, a form of ‘progressive elaboration’, offers a pragmatic approach to managing estimates and achieving the project’s overall objectives. Activities are progressively expanded to greater levels of ‘density’ as more information becomes available:

• Low density is appropriate for work, which is intended to take place 12 months or more in the future. Tasks may be several months in duration.
• Medium density is appropriate for work, which is intended to take place between three and nine months after the schedule date. At this stage the work should be designed in sufficient detail to be allocated to contractors, or subcontractors. Task durations should not exceed two months.
• High-density scheduling is an essential prerequisite for undertaking work. The schedule is prepared with the people doing the work. Task durations should be no more than the update cycle.

A similar approach works for cost estimates, particularly given the cost of an activity is a factor of its duration and resource usage. Importantly, as the density is increased, adjustments to the plan take into account actual performance to date, available resources, work content, difficulty and other factors necessary to achieve the overall schedule objectives.

In summary, a good estimate is realistic based on the information available, supported by a pragmatic assessment of the likely error range. Adding excessive detail when the detailed information is guessed rather than known destroys the value of the estimate in a fog of inaccurate data. The false perception of accuracy created by the heap of data creates false expectations and prevents sensible management of the project based on what is really known.