The Game of Risk

We are living in a time of economic uncertainty. Fine. For some people, that might mean a hunker-down, reduce-risk, stick with what we know mentality. I take a contrarian’s approach: Start the risky projects. If we don’t start the risky projects, how can we discover a breakthrough, reduce waste, or innovate enough to work our way out of the uncertainty?

But don’t start the risky projects without a way to manage the risk. And, don’t start projects that don’t fit with the corporation’s strategy at all. That means starting a project and looking to see what value it and each other project has delivered so far; and reevaluating this project with respect to the other projects underway periodically.

Sounds good, right? There’s a catch. You can’t use a serial approach to the projects, and you have to look at the whole picture for the project portfolio. Welcome to agile and lean project portfolio management.

Deliver Value in Chunks

If you’re accustomed to managing a portfolio of waterfall or phase-gate projects using a serial lifecycle, you’re accustomed to seeing documentation as deliverables early in the project, code later in the project, and finally finished features quite late in the project—if you’re lucky. But to start risky projects, you need to see features—as in running tested features, features you can demo—early in the project. With visible progress, you can evaluate the projects in the portfolio as often as you need to, whether that’s once a month or once a quarter.

So how does a team deliver working product? By using agile approaches: delivering features (or feature bits or feature sets) in a specified time period. The project team can work in a timebox, so you can review the features the team delivered in that time. Or, the team can work incrementally, feature-by-feature, whether that’s in a timebox or not. But no matter how you do it, you have to work on small enough chunks that the project team can finish features, as in fully tested, running, shippable features in—at most—a month. Why a month? Because you need to evaluate how much risk and value this project has and provides.

Developing products incrementally to see their progress is key to taking on the risky projects. That incremental approach allows you to review the project’s progress and know if the risk is too high for now, or if you can manage the risk and complete the project.

If you can’t see product progress, you can’t know how risky this project is and whether to commit to it again or not.

Combine Agile with Lean

Have you ever looked at a project and said, “We’re spending time and money on this? Why?  It’s a waste of our time.” That project received a blind commitment from management, and the folks who make the project portfolio decisions didn’t reevaluate the commitment periodically.

But just looking at the projects that literally waste your time isn’t the only way you can use lean. Think about all the people who are trying to work on two, three, four, or even more projects. No one can work on that many projects and actually make progress on any of them. If you level out the workload by committing to only the projects you have enough people to staff for now, making sure each person is committed to just one project, and then reevaluate the portfolio, you are managing the “production” system (in lean terms), as well as the workload. You’re also keeping an eye on the big picture, and optimizing at the organization level, not the project level.

Another way to use lean is to keep a steady stream of work for the people—not more work than they can accomplish, but a steady stream. That means you need to make decisions about the portfolio before people have to wait for your decision. So, you need to evaluate the portfolio periodically so if a project team completes one project, they can have a celebration and take on the next project on the list.

If you think of lean approaches to the portfolio, you want to make big decisions as late as possible, you want to fully staff all projects you decide to commit to, and you want to stop projects that you can’t commit to or don’t deliver enough value for now.

That means you need to periodically review all the projects in the portfolio, not just the risky ones. When you conduct a portfolio review, you don’t need a dog-and-pony show. You need answers to these questions, so you can take a long-term view of the organization’s strategy and the projects you have:

  1. Does this project still fit with our strategy?
  2. What has the project team finished since the last evaluation meeting?
  3. Where is the team in the product backlog?
  4. What are the team’s obstacles?

Yes, these questions look a lot like the three questions the project team members answer in a standup meeting. It’s worth asking the first question, because the “Should we do this project at all?” is a fundamental question to taking on risky projects and managing the project portfolio. Looking at the team’s progress and obstacles—making sure the team can show their progress and articulate their obstacles—makes it possible for you and the other decision-makers to make thoughtful decisions about the project portfolio.

Should we do this project at all?

The “Should we do this project at all?” question helps people see and discuss the value of each project on its own merits. When you ask that question, people start saying things like, “Oh, it helps these kinds of people” or “prevents this problem” or “allows us to access these customers” and other discussions of the benefits of the project. This is not a discussion of traditional ROI. Especially for risky projects, you don’t have a crystal ball and can’t define the return. Don’t waste your time trying.

Instead, ask the “Should we do this project at all?” question. If the answer is yes, now you can ask other questions such as “Is there still more value in the product backlog for this release? Should we fund this project for the next iteration or time period until we evaluate the portfolio again?”

Commit Means a Full Commitment

If you are ready to fund the project for the next set of deliverables, make sure it’s a full commitment. That means the project has all the people, all the hardware, all the resources it needs, in any form it needs it, to be successful until the next time you evaluate the project portfolio. No multi-tasked people, no waiting on the single DBA who’s working on that other critical project over there. Either fund this project for now, or don’t. But don’t partially commit to a project. That’s no way to make a project successful, risky or not.

Combine Agile with Lean to Manage Risk

When you take an incremental or agile approach to projects, you can see progress, and know whether a particular project is truly risky or not. If you defer big commitments, and look for waste and other ways to apply lean thinking to management, you can start and stop risky projects—any project—and manage the risk. You too can be a contrarian—and succeed.

 

Acknowledgements

I thank Esther Derby and Don Gray for their review.

© 2009 Johanna Rothman

Leave a Reply