As business leaders, we are often called on to make decisions about which option or project we should pursue.  Today, we'll talk about one consideration to help you improve those decisions: sunk costs.

What are Sunk Costs?

When making business decisions, each option you face has associated future costs and associated future revenues.  Typically, you will compare the future revenues to the future costs, and adjust for the timing of the cash flows and for the risks involved.  This provides a comparison of the likely profitability of each option.

Sunk costs are money that you've already spent on one of the options, before making the decision.  Regardless of which option you choose, the money has already been spent.  That money is, for all intents and purposes, gone.  If you choose option A, the money is spent.  If you choose option B, the money is spent.  If you choose to do nothing, the money has still been spent.  The result is that sunk costs should not be considered in your decisions.  Sunk costs do not alter the future costs and revenues of your options, so they should not be included in the analysis.

Let’s say you have two innovation projects.  Project 1 has invested $100K so far.  Project 2 has invested only $10K so far.  You only have the budget to continue with one project.  Which one should you choose?

The answer is: Whichever project has the best future return for the company.  The money spent in the past is irrelevant, because you can't get that money back.  If project 2 has better future returns, but you choose to proceed with project 1, you are essentially "throwing good money after bad".  That is, you are wasting more money on an inferior project, just because you wasted money on it in the past.

Doesn't Everyone Exclude Sunk Costs?

Although excluding sunk costs from your decisions seems to make sense, managers very frequently fall into the trap of continuing a losing investment just because they've already invested in it.  There are a few reasons for this:

Over-optimism: In a study in the Journal of Personality and Social Psychology, researchers found that once an investment has been made, the investor has a stronger belief that the investment will succeed than before they had made the investment.  This has a direct parallel in business.  Many projects slip from month to month to month, because managers repeatedly believe that they are "almost there".  However, if they approached the analysis of future costs, revenues, and risks more objectively, they might instead cancel the project and invest in an opportunity with a better likelihood of success.

Over-optimism also causes some managers to believe in a "sunk cost dilemma".  This is the belief that ignoring sunk costs will lead to an overall bad outcome for the company.  An example: After its first month, a project has over-run its costs and missed its revenue forecasts.  However, those costs are sunk and should be ignored.  Looking at the forecasts, the project still looks promising, so the project proceeds.  After the second month, the project has missed its estimates again and has lost even more money.  But these are sunk costs and are ignored.  The manager, looking forward, only sees a rosy picture, and the project proceeds.  This continues from month to month, until the project completes, showing a large financial loss for the company.

The problem here does not come from ignoring sunk costs.  The problem comes from being over-optimistic about the future outcomes.  After repeatedly missing past forecasts, managers should be that much more diligent about ensuring that future estimates are realistic, instead of getting caught in the trap of repeatedly believing questionable estimates, when past evidence suggests that they are unreliable.  To put it another way, ignore sunk costs, but don't ignore what you've learned.

Personal responsibility: In several studies, including one published in Organizational Behavior and Human Decision Process, researchers found that if a manager feels responsible for the sunk cost, then they are more likely to want to continue that investment, even in the face of better investment options.  This is human nature -- none of us likes admitting that we were wrong or did a poor job.  If you are in this scenario, beware!  Often, how you respond to your mistake is much more important than the mistake itself.  If your project didn't work out, learn to walk away and avoid the same mistakes on the next project.  If you ignore the data and continue a failing investment, you will soon find yourself in an even deeper hole.

Loss aversion: When you walk away from a project with sunk costs, many people feel that they are "wasting" past investments.  Of course, the true waste is continuing to invest in a losing proposition, when that money could be better spent elsewhere.  However, this psychological barrier is a difficult one to overcome.  Some may say, “We’ve spent so much on this already, it would be a shame to throw that away.”  Focus on the future, on how much future money you expect to make for your future expenses.  That will help you avoid turning a loss into a larger loss.

Note also that putting a project on hold doesn't mean your investment is lost.  Frequently, a cancelled project has still created some useful assets, such as intellectual property, that you may be able to reuse in other projects later.

When Should You Consider Sunk Costs?

Although you'll never include sunk costs directly in your analysis, you should make sure you include all the benefits of your past investment in the decision.  Here are some examples:

When the past investment reduces the cost of a future option: Frequently, when sunk costs are involved, you are comparing completing an existing project to implementing a new project from scratch.  Of course, the future cost for the existing project is less than its total cost, because you've already incurred part of the cost of the project.  However, abandoning the project and proceeding with another option may still show the best financial return, especially if the project is slipping and is likely to slip further.

When the past investment creates a barrier to entry against your competitors: Sunk costs can represent a real barrier to entry for your competitors, if competitors would have to make a similar investment to compete with you.  An example of this is when creating an innovative new product.  A product with a barrier to entry means that your market share (that is, your future revenues) could be protected from imitation longer than it would be for a second product which is cheaper for a competitor to copy.  In this scenario, make sure that the revenue forecasts for your options reflect this.  The revenues for the product with the barrier to entry should remain higher for longer, when compared to the product without a barrier to entry.  After all, the faster competitive products appear, the sooner they're likely to start competing on price.  Note that you're still not including the sunk cost itself in the analysis.  Instead, you're including the result of the investment (i.e. higher future revenues) in your analysis.

How to Avoid Hard Decisions

Regardless of how you look at it, walking away from sunk costs is a hard decision to make.  So how can you avoid having to make these hard decisions?

Evaluate the project, not the person: We already discussed that the sense of responsibility makes it difficult to step away from sunk costs.  To make this easier, remember that you are evaluating the project, not the person running the project.  If you focus on the person, they will often become defensive, and promote staying the course, when a change in course is required.  But if you focus on the merits of the options themselves, and take the person out of the equation, it becomes much easier for the people involved to step back and look at the decision objectively.

Ask hard questions early: The best way to avoid having to make hard decisions is to ask hard questions earlier in the project, to make sure the team is learning about its costs, its target market (i.e. future revenues), and is getting its risks under control.  Avoid unrealistic optimism -- Frequently reality check your forecasts, and make sure the team is steadily reducing its risk.  If the team is not getting its risks under control, it might be easier to put the project on hold early in the project, or even to step back to performing feasibility studies, rather than wait until the investment has become significant.  Successful entrepreneurs understand this concept intuitively.  If an idea is not working out, they move on to the next one, before they've invested too much in it.

Iterate rapidly and inexpensively: When your software activities are implemented iteratively, and each iteration is rapid and inexpensive, then you have built-in milestones where the project can be evaluated objectively.  If you decide that a project does have to be cancelled or changed significantly, then you've minimized your past investment, and the team has a point where they can change course quickly and easily.

Lastly

I've touched on several topics in recent weeks.  Which ones are most important to you?  Innovation, culture, risk, and software management techniques are each large topics, and I want to make sure that you're getting value out of these articles.  Let me know...

Of course, if you have any questions, or if you would like more information on how to implement these or other software development processes in your organization, please feel free to contact me at Charles@CharlesConway.com.

If you know of someone who may find this article of interest, please forward it on!

Good luck!
Charles