In most organizations adopting Agile methods, the techniques used for program and portfolio management are still predictive and “waterfall”: yearly budgeting cycles, capacity planning, and heavily matrixed resource management. It’s no surprise then that despite adopting Agile methods for their projects, many organizations have yet to exploit their full benefits at the portfolio level.

Put another way, Agile projects are constrained because their portfolios are clogged with the debris of failing projects, with slow moving projects that delay more critical initiatives, and with the inability to properly staff projects to get them moving at a rate that their business customers would appreciate. How can the project portfolio be fixed so that business value can flow and projects can be completed faster?

Program and portfolio managers have the charter to apply systematic management techniques to collections of projects and deliver business value in line with organizational strategy. In my previous posts on the subject, I laid out principles and practices for a Lean-Agile PMO that use lean principles to accelerate Agile project delivery.

To help you get started, here are 5 steps that you can use to make your project portfolio flow:

1. Terminate Sick Projects. In every organization, there are sick projects. These are projects that fail to deliver value: missing deadlines, not meeting business expectations, and undergoing long and painful “illnesses.” It is reasonable to assume that the majority of investment sunk into sick projects ends up wasted—returning very little value to the business. Consider the time spent by team members on those projects. If the portfolio was purged before these projects, and the project inventory thereby reduced, the team members’ effort could be redirected to more productive projects that in turn would finish faster. To begin clearing the debris in your portfolio, don’t avoid the challenge—terminate sick projects as soon as you can.
2. Break Large Projects in Smaller Projects/Increments. In Agile parlance, a minimum marketable feature (MMF) represents a component of intrinsic marketable value. By categorizing and aggregating product features into MMFs, you can recognize and exploit the fact that sets of product features can have value to end users, even if the product is incomplete. Instead of carrying massive project inventory to deliver product functionality in one big bang, deliver products in increments of MMFs and improve lead time as a direct result. To reduce the undesirable large Work in Process (WIP) created by large projects, break them into smaller projects (or at least deliver smaller product increments) organized around MMFs.
3. Manage the On-Ramp. Is your organization is starting more projects than it’s completing? This might a sign of improper project kickoff discipline that could lead to future project delays. Obviously, not all projects are alike and it is possible to end one large project and correspondingly start up several new smaller ones in its place, but this metric can still serve as a potential warning device and leading indicator of delays to come. To manage the project “on-ramp,” create a lightweight but disciplined project prioritization process to kickoff projects, and follow it stringently.
4. Create Stable Teams. To maximize project throughput, an organization should only start as many projects as available teams. Instead of sinking costs into forming teams for projects only to break them up on completion, smart organizations create stable teams as high-performance units focused on one project at a time. Each team is as fully cross-functional as practical, containing business analysts, development leads/designers, developers, testers, and perhaps a tools support person. An organization may also have separate dedicated support teams. You can then pull projects from a prioritized project backlog of selected projects and allocated to the appropriate team. The team focuses on this project alone to the extent possible and in working closely with the business sponsor, should be able to finish it way faster than in traditional environments. When the team completes the project, puts the system into production, and hands the maintenance over to the maintenance team, the team will then be available to start the next project. Create stable teams by dedicating core team members at least 80% to the project at hand.
5. Track and Control the Flow of In-Flight Projects. Projects are typically measured along the triple constraints: scope, schedule and resources. Using these as sole measures of progress is problematic: Because scope is often variable, schedules are usually assessed through unreliable artifact-based milestones.
The flow of business value in your project portfolio from Initiation, where ideas originate, through Deployment, when value is realized, should be smooth and steady. Variability – so desirable in Agile product exploration – is not a good thing in the project portfolio.

Defining metrics based upon the value propositions of product MMFs can provide much better indicators of progress. Rather than focusing on project constraints, milestones can measure the realization of product goals. When projects fail to live up to their business cases, as assessed by frequent and thorough evaluations (preferably with real target users), they can be retargeted or cleared from the portfolio. Institute a portfolio prioritization and control process that is based on the delivery of business value, and use it to relentlessly reevaluate projects in-flight and keep WIP in the form of in-flight projects to a minimum.

Get your portfolio flowing, and get started on the journey to low variability, high throughput and high value by applying the five steps above.