Modis Whitepaper | Are You Making These Five Common Estimation Mistakes?

Are You Making These Five Common Estimation Mistakes?

Louis Mierowsky Posted 20 April 2015

While ‘non-estimators’ expect estimation to be an exact science, there are a large number of easily-made mistakes, that can have a major impact on both project cost and schedule.

Make sure you aren’t heading into a project blow-out by avoiding the most common ones.

1. Understand the difference between estimates and targets

Targets are an ‘ambit’ position often used to frame tenders.

For example, a client will usually go to market with an explicit or implicit budget and a set of requirements. While a useful starting point, an estimate tests the target through careful analysis of client and experiential data. Commitments to clients should be on the basis of estimates not expectations.

2. Identify dependencies and explain their potential to affect estimates

A failure to clearly elaborate dependencies (such as the linked agenda of another business area or the requirement for prerequisite infrastructure) will affect your project schedule and resource optimisation (and hence, commercial return).

That is, you may have planned for a project to run at 90-100% utilisation across a 12 month period but the effect of an unforeseen dependency may drive this down to 70-80% across an 18 month period. If risk, contingency and margin have been calculated on the basis of the former then, potentially, it risks the commerciality of the project (especially if it is fixed price). From a client perspective, this completeness of view is always appreciated and preferred to being surprised mid-project when they have little if any capacity to take remedial action.

3. Balance resources across the project schedule

Estimates typically build from experientially derived effort standards for known activities and allotted to the relevant competency, then converted to FTEs.

Initially, how the number of FTEs converts to team size is a product of a ‘target’ deadline and/or a roughly sequential application of the development lifecycle e.g. concept, blueprint, build & deliver. While this gives a rough resources ‘fit’, it’s usually not the optimal use of resources. The team may require slimming up or down, incorporating parallel streams and/or staggering activity to account for dependencies. Unless this balancing or ‘smoothing’ occurs, a client can view project effort as under or overdone which can adversely affect client confidence when a genuine resource constraint is raised through change request. Right-sizing the team from the beginning ensures optimal productivity and client satisfaction.

4. Recognise the things you can’t change

Projects will always have a mix of variable and fixed activities for example the competency mix of a project team that can be fluid compared to data migration, report construction or development of a new ESB service.

This may be able to be improved slightly with software tools but there is usually a fixed kernel of effort below which these things cannot be safely delivered. This is something a client would not usually be expected to know so explain this either in your estimation or early in the project. Again, this is invariably appreciated by clients who can then act early enough to manage their own stakeholders’ expectations.

5. Understand risk and contingency

Good estimation assesses the level of risk attaching to a project.

In a tender situation, this is determined by a variety of factors including governance, client maturity, third party involvement, whether price has to be fixed and the level of requirements definition. The higher the risk, the greater the contingency i.e. the effort/cost margin added to price to deal with the uncertainty deriving from risk. Not recognising the correct level of risk, especially in a fixed price project, will lead to an incorrect contingency factor being applied to an estimate.

If the contingency is too high, then the organisation may not win the work. If the contingency is too low, then the organisation will lose money on the project. It is often useful to call this out in fixed price projects (if it is not specifically requested). Again, risks are not always known to clients and where raised (and accepted) can result in mitigation and an agreed reduction in price.

Summary

There are many other factors that can impact project success such as scope creep, optimistic schedules, short-changed quality assurances, excessive multi-tasking, the abandonment of planning under pressure and wishful thinking!