How do you handle sunk costs of projects at project portfolio reviews?
I'd like to hear from people who periodically review their project portfolio using financial metrics such as NPV, DCF, or maybe a ratio like projected (Gross Income) divided by (development cost). Of course, the first time you review the project you will be using estimated values for costs and benefits. However, at subsequent reviews you would expect to have better estimates of potential benefits. Liekwise, some of the costs will have actually been incurred (sunk). Do you recognize the sunk costs in your calculations at subsequent project portfolio reviews?
Let's say you used (Projected Gross Income) / (Projected Total Development Cost) to initially approve the project. Would you use (Projected Gross Income) / (Projected Total Development Cost) or would you use (Projected Gross Income) / (Remaining Development Cost) at subsequent reviews? Why?
To better understand the implications of the question, see my blog post about when to kill a project at http://blog.clarosys.com.
Good Answers (1)
Peter F.
Product Development Management Consultant
Best Answers in: Business Development (1), Project Management (1), Market Research and Definition (1), Industrial Design (1), Product Design (1)
Greg,
It depends upon the purpose of your review.
If you are making continue/don’t continue decisions relative to which projects in your portfolio get limited funding, the decisions should be based on today’s estimates of the expected benefits vs. the remaining costs to complete for each of the projects in the portfolio. If Gross Income is the metric you prefer to use for benefit, all else being equal, you would compare Today’s Estimate of Projected Gross Income/Today’s Estimate of Remaining Development Cost. Your decisions today are independent of your decision history.
If the focus of your review concerns evaluation of your ability to make valid initial portfolio decisions, you should compare this same ratio to your initial estimate of Projected Gross Income/Projected Total Development Cost. Your decision-making performance depends directly on your decision history.
Pete Frickland
Concurrent Product Development, LLC
More Answers (3)
David G.
Project consultant at KPMG
Best Answers in: Project Management (11), Organizational Development (2), Planning (1), Enterprise Software (1)
Greg,
I have run an Enterprise PMO for some time, and stick to a simple rule - no decision is to be made on the basis of sunk cost. I think Pete Frickland's answer is clear. The issue about whether to continue or not is about how much benefit is available from investing the remainder of the budget. The rest is gone. It is a difficult position to be in and very tempting for executives to kill a project because it is way over budget, but the problem is in the past and cannot be influenced. Yes, the ROI will be awful and if you could go back you wouldn't do it again, but the decision of the day must be about the future and whether benefits are achieveable by investing the funds required.
Cheers
David K.
Technical Account Manager at NBC Universal
Best Answers in: Quality Management and Standards (1), Branding (1), Small Business (1)
A good manager knows when to stop throwing good money after bad. That said, sunk costs are just that and should not influence a go/no-go decision later on in the process. The best way to relate this is poker.
If you had a good hand to begin with, you've essentially okay'd the project and made a decision as to what you're willing to risk to get to the discovery phase. This is often the most blind situation and you're paying to see what can possibly come out of this.
Once the flop comes out (next round of betting), you decide if your hand is still good or not based on the possible benefits (current pot). If the costs go up (someone raised), you decide if the potential benefit is worth the cost to continue based on your ability to win. You weight this benefit against what you will need to put in, not what you already have as you have no way of getting that back other than to win the pot.
Most managers will blindly spend more money to fix a problem with a project saying that just a few more dollars will reap the same benefits as before. This is the equivilant of bluffing in poker. This is a losing proposition as the ability to win is not considered.
Any tollgate review MUST consider the ability to complete the project successfully on top of the ROCI (Return on Continued Investment).
Another way to look at it: To fix your used car or to junk it (or sell it for what its worth)... After a certain point, the repair costs (development cost) per mile (benefit in this case) compared to alternatives (buying/leasing/renting/public transportation) make it no longer feasible to pour more money into it.
You also have to consider whether some of the sunk cost can be recovered if you were to quit now and if that benefits you more than continuing with the project. Just like selling the car as a fixer, you probably have equipment purchased that can be sold or used for other projects.
For your project, your review also needs to include whether new developments in the industry or information you gathered shows a different approach that would be more beneficial to the company. Examples: Buy a hybrid for fuel economy (new technology that allows for lower operating expense) or buy the same car you have in good condition for less than what it would cost you to fix the car you have (buy vs. build, someone developed what you're trying to develop already and its cheaper to buy off the shelf).
Basically, use the amount you need to continue to weigh against the benefits expected and also consider the expectations of success (within budget) and any better alternatives. There are many more factors and ways to analyze project feasibility, but I could write a novel about that...
Jim P.
Senior Business and Technology Executive specializing in Advanced S&OP, Business Intelligence, and ERP Implementation
Best Answers in: Inventory Management (3), Corporate Governance (1), Change Management (1), Manufacturing (1), Supply Chain Management (1), Enterprise Software (1), Information Storage (1)
I agree with the answers posted before this -- if the primary purpose of the review is a future investment decision, Projected Remaining Development Cost is the relevant number. Having said that, if the project should be stopped, I would want to know whether the problem was that the Projected Gross Income has changed, or that the Earned Value of the Project to date is very low, based upon the plan. Also, if the decision is at all close, I would want to understand the impact on the morale of the organization and commitments made to various internal and external entities as part of the evaluation.
Now, if the purpose of the review is to identify which projects have been successful, gather best practices or establish baselines, the Projected Total Development Cost is more relevant.
