Avoiding sunk cost syndrome on your project

Patrick Weaver
February 23, 2015

One of the hardest things to do is to stop wasting money and resources on a losing proposition. The fact you have spent several months and thousands of dollars on a mission to accomplish something should not influence your thinking about expending more time and money—but it frequently does.

The past is past, and money spent is gone forever; the costs and commitments have been ‘sunk’ and cannot be recovered. Making decisions about the future based on these ‘sunk costs’ simply distorts the decision and frequently leads to really bad outcomes.

There are basically three scenarios where sunk costs influence decision making:

There is very little prospect of ever achieving any return on the investment: the sensible option is to stop wasting resources and kill the endeavour regardless of the amount of cost and time already expended. Wasting more resources does not mitigate the waste to date.

  1. The return on investment will be less than the additional resources that need to be expended to achieve the return: again the only sensible decision is to kill the endeavour regardless of the amount of cost and time already expended. Wasting more resources does not mitigate the waste to date.
  2. The return on investment will be more than the additional resources that need to be expended to achieve the return, but will be insufficient to cover the overall cost of the work: in this situation, expending the additional resources to reduce the overall loss may be sensible.
  3. The change from option 2 to option 3 above is time related. Consider a project that has suffered a major setback since initiation and the expected benefit is now only $10 million from a total project cost of $20 million: with perfect foresight, we know that if the project is completed, over its life there will be a net loss of $10 million.

If this situation is identified at the 25% stage, the expenditure to date will be $5 million, cancelling the project results in a net loss of only $5 million (the sunk costs). Spending the other $15 million will allow the benefit of $10 million to be realised, but increases the overall loss to $10 million. The sensible option is to cancel the project, and minimise the overall loss.

If the situation is identified at the 50% stage, the expenditure to date will be $10 million, cancelling the project results in a net loss of $10 million (the sunk costs). Spending the other $10 million will allow the benefit of $10 million to be realised, but the overall loss remains at $10 million. The sensible option is to cancel the project as the extra capital and resources would be better used on a project with a positive return on investment.

If the situation is identified at the 75% stage, the expenditure to date will be $15 million, cancelling the project results in a net loss of $15 million (the sunk costs). Spending the other $5 million will allow the benefit of $10 million to be realised, which will reduce the overall loss to $10 million. The sensible option is to keep going and spend the money to reduce the overall loss from $15 million if the project is cancelled now, to $10 million if it is completed and the benefit realised.

The trick is not to think about cost

The key factor affecting the above scenarios is that in each decision, if based on achieving the best overall outcome for the organisation, the planned project costs are not given or discussed—they are completely irrelevant. The fact the project costs have increased from $15 million to $20 million should make absolutely no difference to the stop/go decision.

The major problem with this scenario is that the actual costs are known for certain; future costs and benefits are estimates and will always be uncertain, which is to say there are risks involved. This is compounded by time and control factors. The further into the future you try to predict costs or benefits, the greater the uncertainty.

Similarly, most organisations have significantly better understanding and control over their costs than over future unrealised benefits. These uncertainties should be part of the decision making process and adjustments made to the future estimates to allow for the uncertainty by using benefits from the lower end of the expected range and costs from the higher end.

From a rational perspective, it is highly desirable to fail early and fail cheaply as in the 25% scenario above. The problem is that at this stage of the project’s lifecycle, the ability to ‘precisely foresee’ the final total cost and benefits is significantly less than later in the project.

The information available for decision making evolves over time and may change. The ‘kill’ decision needs good information, careful analysis, clear thinking and a degree of courage to act on the available data. Unfortunately this type of decision is rarely based on rational thinking, emotions and innate biases have a major effect.

(This is a companion piece to How not to kill a project)

Author avatar
Patrick Weaver
Patrick Weaver is the managing director of Mosaic Project Services and the business manager of Stakeholder Management Pty Ltd. He has been a member of both PMI and AIPM since 1986 and is a member of the Asia Pacific Forum of the Chartered Institute of Building. In addition to his work on ISO 21500, he has contributed to a range of standards developments with PMI, CIOB and AIPM.
Read more