Estimates are required, to varying degrees of accuracy, at all stages of the project life cycle. Before an organisation can make an investment decision, an estimate needs to establish how much capital investment is required. An estimate is likely to be needed before any shareholders or stakeholders authorize funding for investment. Further along the project life cycle an estimate is required for tendering and bidding processes and throughout the life cycle the estimate is a control; a document that project stakeholders can reference progress against.
Estimating accurately can be a difficult process but is one of the most critical components of a successful project. The classic challenge for all project managers is to complete their projects on time and within budget, yet projects in every industry often exceed their original projections. The key to producing a good estimate is quality of information. As far as possible each item, resource, indirect cost, overhead, logistical element, paper clip or paper cup! needs to be known if it is to be included in the cost of the project.
Throughout the Project life cycle the accuracy of an estimate will change. Estimates are, after all, only quantified approximations of project costs, durations and resources. Estimates however are vital to any project, without them you cannot put together a schedule, find out what your resource needs are or even draw up a budget. An important factor to keep in mind is that an estimate can never be 100 per cent accurate (unless you are extremely lucky).
The start of a project is the hardest time to estimate, due to the level uncertainty over key factors like what is to be done, when, by whom, etc. At this stage an estimate’s range is likely to be wide, compared to the latter stages of the project when the detail is more clearly known.
There are four main methods of estimating:
Bottom-up estimating (analytical)
Bottom-up estimating requires the identification of each discrete activity required to complete the project. For each activity the resources (labour, materials, equipment or financial) and the elapsed time required to complete the activity are estimated. During this process it is important to allow for normal staffing overheads (leave, training, sickness etc). The activity totals are then added up to give a detailed estimate for the project. This approach is usually used when the fine detail of the programme and/or its component projects (work packages in the case of projects) are well defined.
Comparative estimating (analogous)
Comparative estimating involves comparing current or completed tasks or projects for which you have some measures of time and resources required. This method is based on past experience rather than opinion, but it is only useful if the analogy is valid. The comparative method provides a firm basis for estimating if information is available, this information would then need to be scaled up or down to meet the needs of the project which is being estimated.
Parametric estimating (statistical modelling)
Parametric estimating uses defined parameters by which a project can be measured, such as time or cost involved to build a specific project deliverable. This process can be repeated for a number of different deliverables, multiplied by the number of each of the parameters required to fulfil the project requirements. A reasonable amount of robust data is required in order to make this an easily accessible estimating technique.
Top-down estimating
The most accurate way to estimate is usually to build a Work Breakdown Structure (WBS)and to estimate all of the lowest level, individual work components. However this bottom up approach is the most time consuming. It is also not appropriate for initial estimating that this level of effort is expended at the early stages as a rough estimate may be all that is required to give the project a go ahead or not.
A Top-down approach can therefore be utilised to gain as much estimating confidence as possible. Top-down techniques include the following:
- Previous history – this is using some of the comparative techniques discussed earlier. This is by far the best way to estimate work but is obviously dependent on delivering similar projects in the past and also relies on keeping track of actual effort, hours and costs from previous projects.
- Parametric estimating can also play a major part in the top-down approach where we can use previous similar projects and multiply up or down costs and duration to reflect our new project.
- Partial Work Breakdown Structure (WBS) – this approach involves building a traditional WBS as in the bottom-up approach but you take it down one or two levels.
- Expert opinion – In most cases an internal or external expert will be required to help estimating the work to be completed. Often benchmarking with other companies in the industry is useful where leverage from their experience can be utilized.
- Estimate in phases – this is often called rolling wave planning or progressive elaboration. One of the most difficult aspects of estimating projects is not knowing exactly what work will be needed in the future. To reduce the level of uncertainty, you can break the work into a series of smaller areas of work, only estimating accurately the most current work with a vague or best guess estimate for the remaining work.
It should be noted that an estimate should not be treated as an exact value – it is an important part of the estimating process to ensure the recipient of the estimate appreciates the range of potential outcomes. A robust
Risk Management process should be in place to allow for contingency funding within the estimate.
