Point Play

With effective project portfolio management, everyone knows which project is #1, which is #2, which is #3, and so on down the line. Then you commit to those projects in order, by making sure they have teams who can do the work required, that the projects have any other resources they need until you reevaluate the project portfolio again.

But, sometimes, it feels as if those of you on a leadership team—the people making the portfolio decisions—really can’t decide between two projects. You’ve discussed the benefits of each project until you’re talked out, and you still can’t decide. You have a few options to help you decide: ranking with points, try an iteration for each project, and alternating iterations to see where the “gotcha’s” are in each project.

Use Points As Your Ranking Method

You may have tried to discuss the relative merits of each project or compared them against each other. Often, that works. But more often, it’s still difficult to decide which project is first, which is second, and so on. That’s when it’s worth using points as your ranking method for projects. Points help you see the relative business value and ignite discussion about the relative value of each project.

When you rank with points, you're separating business value from funding. The number of points you assign to a project is a representation of its value to the organization, not the funding you will provide. This separation of value from funding works in a similar way that separating project sizing from duration helps project staff estimate better.

To rank with points, take a large number of points. If you’re trying to decide among five projects, start with 100,000 points. Now, assign a unique number of points to each project. Here’s how one leadership team started assigning points:

Project Value Points
Project 1 20,000
Project 2 19,999
Project 3 19,998
Project 4 19,997
Project 5 19,996

Table 1: Initial Point Assignment

Not much difference among these projects is there? There is a difference, but the leadership team decided the close ranking meant that all the projects were essentially the same. If you find yourself in a similar position, consider asking some what-if questions:

What if we doubled the number of available points? Would the ranking still hold?

When this leadership team doubled the number of points to 200,000, they ranked the projects this way:

Project Value Points
Project 1 60,000
Project 2 55,000
Project 3 28,000
Project 4 25,000
Project 5 24,000

Table 2: Ranking After Using Twice the Original Points

The relative ranking still held. But notice the difference between the first two projects and the last three.

As the leadership team started talking about how much more valuable some projects were than others. Now, they have two clear winners: Project 1 and Project 2. By using more points, it’s easier to differentiate the projects’ value. With more points, the leadership team had more complete discussions about the relative value of each project, so they were able to discuss why the top two projects were more valuable and the other three were somewhat less valuable.

What if we multiplied the total number of available points by 10?

Sometimes, doubling the points doesn’t work, so it’s worth using ten times the original number of points. There’s nothing magical about the number 10—it allows people to differentiate each project’s value more clearly.

Another leadership team had only four projects to decide among, but they could only staff two projects.

Project Value Points
Project 1 2500
Project 2 2499
Project 3 2498
Project 4 2497

Table 3: Initial Ranking with 10,000 points

This team discussed for a while, and then decided to use 100,000 points. They started with this:

Project Value Points
Project 1 75,000
Project 2 25,000
Project 3
Project 4

Table 4: Start ranking with 100,000 points

That’s a surprising change for a team that had four projects so close together initially. Because they gave themselves more points to use for each project, they started discussing why they were considering projects 1 and 2. One of them explained the value of Project 1 to the rest of leadership team, and each team member realized Project 1 was not just the first priority, it could even be the only priority if they needed more people on it; that the benefit of doing Project1 outweighed everything else the organization was doing.

What if I removed one of the projects? Would the ranking still hold?

Sometimes, the projects appear to be linked. If you remove a project, how would that affect the rankings of the other projects? If all the projects are dependent on one project, that project might be your top-ranked project, and you don’t have to worry about all the other projects until that one is complete. If you remove a project and there is no change on the ranking, maybe you have the right ranking.

Maybe you’ve tried all of these alternatives but you’re not making progress. Your leadership team still thinks you have more than one #1 project. In that case, try an iteration’s worth of work for each project, at the same time using two different teams, and reevaluate the projects.

You may have other what-if questions that work for your leadership team or your environment. The key is to instigate discussion around what a valuable project is in your organization.

Try An Iteration for Each Project

Sometimes, the way to instigate discussion about what “valuable” means is to request a team does an iteration’s worth of work on a project. Once they’ve completed an iteration of work, the leadership team can re-evaluate how valuable a project is.

This approach is useful especially when you want to understand how much a project might cost. If you have teams already comfortable with each other, you might be able to predict a project’s worth of velocity based on just one iteration.

If you are new to agile, or if you have a new team, give the team three iterations to know how to work together as a team, and use that third iteration’s velocity as a predictor for the rest of the project.

Once the teams have reported back to you on their iteration’s progress (or on the three-iteration’s progress), the leadership team can re-rank the projects based on each team’s progress and how much more there is to do.

Never Compare Two Team’s Velocities

As the teams work, you want to make sure you are comparing each team to its prediction, not to each other. You can’t compare teams and their velocities. Each team is unique, so each team will estimate differently. One team might think they can accomplish 40 story-points worth of work in one iteration. Another team might estimate the same work as 48 story-points, or as 65 story-points. You can only look at what one team predicts, and accomplishes out of their backlog. You can look at another team, and see how much more valuable one backlog is against another. But you can’t compare teams. Don’t try, or your teams will game the system and you won’t have the data you need to make good decisions.

Alternate Iterations

Maybe you have only one team and a huge number of potential projects. And, maybe you’ve managed to get down to two #1 projects. At that point, it might be worth your time to have the team alternate short iterations (no more than two weeks), working on just one project during each iteration, to see which project becomes more valuable.

Iteration 1: Project 1

Iteration 2: Project 2

Iteration 3: Project 1

Iteration 4: Project 2

Do not ask the team to work on two projects in one iteration. If you introduce multitasking in an iteration, you can reduce the velocity of a team to almost zero. You can’t know which one they will spend the most time on. But you can ask them to alternate their iterations on each project. Do this for up to six iterations. At that point, its time to make a decision about which project is more valuable than the other.

Summary

Because it’s difficult to make project portfolio decisions, using a ranking method that helps people discuss why they want to staff one project over another helps the leadership team make a better decision. That’s why points help.

In addition, gathering data about each project’s progress and risk, which you can obtain after a small number of iterations helps the leadership team decide. And, if you really can’t decide, you can alternate iterations on two projects for a few iterations, so you can gather data and decide.

Whatever you do, decide. Do not allow anyone in the organization to think you have two #1 projects. That will guarantee multitasking and slow everything down. Even if you have to change your mind later, choose one #1 project. You and your project teams will be happy you did.

© 2009 Johanna Rothman. This article was originally published on Gantthead.com

One Reply to “Point Play”

  1. Great ideas! Another way to prioritize is by using NPV (Net Present Value). This project is going to take X amount of time and Cost Y dollars, and the end we will make (or save) Z dollars. Plug that into excel to get an NPV, then use that to help weight the ranking.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.