• Seilevel Team

    Here’s the Team

    Welcome to our Seilevel Photo Op.

Using software requirements to determine whether to continue with a project

Can a software requirements model help you determine if your project should be cancelled? The short answer is “yes.”

By using Seilevel’s Business Objectives Model, the problems your project is supposed to solve and the project objectives are clearly defined. If the project is not currently meeting those objectives or solving those problems, a change of course is necessary, and maybe the best decision is to end the project.

Breaking news: The USAF recently made such a decision – after spending $1 Billion. Read details here. The USAF may want to take a look at the requirements definition process and the models used in their “lessons learned” process.

Surprisingly, sometimes deciding to lose money on software projects is the best decision. Nobody wants to lose money, especially on software projects. Even worse, nobody wants to make a decision that guarantees losing money. If you have an existing software project that has already consumed millions of dollars and thousands of hours, the last thing you want to do is cancel that project. That would lock in your losses. That would turn the hope for future benefits into a huge financial loss. 

Who would make that decision?
A savvy manager, that’s who; one whose decision-making is not distorted by sunk costs.

Sunk costs are investments in time and money that have already been spent and can’t be recovered. No matter what decisions you make going forward, the sunk costs are gone. The problem with sunk costs is that they build up a bias toward continuing projects that shouldn’t be continued. If you’ve already spent $10 million dollars and three years on a project, there is a great temptation to continue investing money in the project, so as not to lose the money already invested.

But unfortunately, the money is already lost. All the money spent is a sunk cost that is gone and isn’t coming back. The best thing you can do is ignore the sunk costs. Pretend they don’t exist. This is very difficult to do psychologically. It is hard to forget about the fact that you lost millions of dollars. Yet this is exactly what you need to do to in order to choose the best path forward.

You look ahead at the costs and benefits of the project in the future, and don’t look back at the past. Even if you’ve already invested $10 million and all that is required to launch the project is an additional $10,000, that $10,000 must be weighed against the benefits of the project, not the $10 million.

As noted earlier, a good tool to determine whether to continue with a project or not is the Business Objectives Model (which is described in detail in the book Visual Models for Software Requirements). Pay special attention to the success metrics part of the model. These metrics tell you what measurable benefits you’ll see if the project is successful. If these benefits don’t add up to the future costs you’ll incur, then it probably doesn’t make sense to invest any further in the project.

There is a flip side to sunk costs: they also make it easier to continue with projects. That’s because, in your cost-benefit analysis, you don’t need to consider any of the previous investments. If you’ve already spent $10 million, but only need to spend $10,000 to finish the project, your benefits only need to clear the $10,000 hurdle, not the $10 million. All your past investments should not be a drag on your future decision making, because past spending is gone regardless of your decision, so it should not factor into your decision at all.

The toughest part of ending projects is that you are deciding to lose money. Your decision is what will guarantee the loss. Deciding to lock in those losses is never easy, but is often the best decision for the future. It takes courageous leadership to admit that a project has failed or is no longer needed, but that’s the only thing that prevents your company’s bottom line from being haunted by the ghosts of past decisions and projects that won’t ever turn a profit.

Leave a Reply

Your email address will not be published. Required fields are marked *