Here's how to address a common problem of inaccurate numbers. .

• The dilemma

  During the business planning process, the estimation is sometimes not much better than best-guess. After all, the project may not start for a year or more, or the work may not even be funded. Closer to delivery the effort may well appear to be double what was estimated. However, the client is still taking the line that it needs to done close to the original estimate.
  That may not make sense. The driver may the need, or want, to get the project done in order to meet objectives.

• Mentor advice

  This is an uncomfortable position to be in. Pressure from a client to commit to a project under estimates that may not be accurate. There is a natural tendency to want to please the client and a desire to do the work. Let's think about what could happen.
  If the team begins the project at this point, and it does take twice the original estimate, there will be trouble, the sponsor who funded the project will be mad at everyone involved, and the project will likely be cancelled. If the estimate is only 10 per cent more than the client's figures, then probably the incremental funding would be available. If the estimate was 20 per cent higher, the client might be able to reduce scope enough for the project to continue. However, if the new estimates are correct and the project will likely cost double the sponsor's anticipated cost, the project cannot go forward.

  It would be wrong to start the project when it's unclear just how much more the project will run compared to the rosier estimate. This causes a difficult situation to occur. The higher estimate may cause the project to be cancelled. The client doesn't want that to happen, so the pressure is to say nothing and to pretend that the original estimate is still valid. This puts the burden on those delivering the project not the client. Here are some actions that can be taken to improve the situation:

  • Discuss all the estimating assumptions with the client to make sure there is no misunderstanding about the deliverables or the project scope. There's no need to raise a red flag based on incorrect assumptions.
  • Double-check the estimates. Perhaps a second estimating technique could be used to validate that the work is as much as is thought. There is no point in creating panic people if the estimating logic is flawed.
  • Have another honest discussion with the client and ask for ideas that would help reduce the number. Reducing functionality is the place to start. With less functionality, the costs should decrease. For example, perhaps some of the deliverable can be manual rather than be automated. Perhaps the project will gain most of the value needed by restricting the deliverables. If the scope or the requirements are reduced, this should clearly be communicated to the sponsor and included in the project definition to avoid any surprises.

If the estimates are valid, and the project scope or requirements cannot be changed, there is no choice but to stick with the higher estimate. It may be that the value will still be there, and the company will fund the project at the higher level. However, if the project no longer makes business sense, it is better to find that out now, rather than halfway through the project.

The business plan should have detailed the financials. This being the case, a detailed examination should explore the rational and money. This will highlight the problems and this should be brought to the attention of the stakeholders and signatories of the project.
The political ramifications do need to be addressed and that should be done in the light of day. When accepting the job a distinct line of reporting that includes senior management needs to be in place.

Although it's best not to showcase flaws, it is best to document everything including an inaccurate budget. If the sponsors are not willing to adjust something to accommodate a bad budget, resign and indicate why. This is simply making the best of a bad job and simply being blamed for less.

1. Ask for more time rather than implement systems with unresolved quality problems
2. You must build a PMO that makes the most sense to your organisation
3. Defining and supporting project management methodology
4. The PMO should perform audits and assessments to validate progress
5. Leverage the PMO to consolidate project status and metrics