Back to Blogs

[Podcast] The Virtues of Lean Portfolio Management

  • Publish Date: Posted 26 days ago
  • Author: MATRIX Agile Practice

[Podcast] ​The Virtues of Lean Portfolio Management

Listen to the latest episode of The Agile Reformists podcast from MATRIX Agile Coach Margaret Korpak below, or keep reading for a summary. Subscribe to The Agile Reformists podcast on Spotify to keep up with our episodes released every other Tuesday.

It is still common for Agile practitioners to view themselves from an Agile mindset lens, but to view budgeting for projects in a traditional way centered around deliverables.

Lean Portfolio Management (LPM) is the idea to transition budgeting from project to product mindset. The focus is more on products and the people needing to do work, rather than funding projects.

How does LPM work?

You are really focused on a bucket of money and you understand the value of priority of work you need to get it done.

Doing so, you are able to be more focused instead of having money in one project for one project and move it around. If priorities change, it’s not a big deal because you are sharing money around rather than people.

Funding is mapped to themes, epics and value streams, not individual deliverables.

Capx vs opx mindset

For budgeting, with a traditional model it is very easy to understand capx versus opx investments. With LPM, instead of capx assignment to specific projects, you must think of capx as new development and opx as operations or maintenance work.

Shifting this mindset is tough at first, but once you do, it is much easier to keep track of capitalized expenses. You don’t have to worry about whether work is capitalized or not, since everything in new development is capitalized.

Analysis and validation of new ideas would fall into opx until it is determined that they would come into fruition.

Alleviating transitions

LPM seems like something companies would want to embrace but shifting from project based to product-based funding and outlook to identifying value stream is difficult.

Here are some steps that help alleviate the transition:

  • Work with the finance team to explain difference from project to product. How it makes sense to manage moving parts and pieces from a money perspective.

  • Have conversations.

  • Allow time to brainstorm and hear concerns from customers.

  • Allow finance teams to become acquainted with it over time. After they get on board, transition this throughout the organization with “chunks” of team at a time, to reduce the culture shock. 

New roles

Put some roles in place around governance and make sure compliance is met, especially in industries where compliance is important, and any regulatory issues are met.

Work at portfolio level to make sure the right things are being funded. Now because you might not have as many checks and balances at the project level, you need to do initial reviews and ensure value that is defined is being met.

Ensuring alignment

LPM focuses on higher strategic themes coming down from enterprise level and how they will break down into epics. You are looking at bigger bodies of work and breaking them down to program level and into features, building more defined value and how these solutions flow down to team level.

Instead of team-up and focusing on projects, you focus on flow of value and value delivery and how to do it with people and use of features and epics.

Quick feedback loops

One fundamental aspect of Agility is short feedback loops. Scrum teams have this in the form of end of sprint demos.

So, how can clients maintain quick feedback loops at portfolio level when it may take months or years to realize these?

At portfolio level, you are breaking work down to epics and corresponding features and user stories to the features. You will be tracking work through portfolio Kanban boards and you will have very defined definitions of what a Minimal Viable Product (MVP) is for epics.

You have columns on Kanban work for waiting to be approved, and work in progress and as features in user stories. There are still demos by the team that will be continuously completed. You take those demos and objectively look at MVPs of epics, and still maintain fast feedback loops to make sure value is being achieved at portfolio level.

By the way, MVP is the minimal amount of work that can go out the door that will still deliver value to them. We focus on the MVP because it will get faster feedback from customers about what will work or not and get it out the door.

Eliminating waste

You have those defined MVPS for epics. You might have a huge scope defined, but as the work gets done, you demo it and compare it to value you want to deliver. You might determine that something that was valuable is no longer valuable. and you can reduce it.

With projects, you don’t know if work is done until the end of project, even in an Agile framework, because there are so many moving pieces.

When you are focused on value with LPM, it is easier to get people together and determine waste early on. Thus, having portfolio alignment allows you to identify and eliminate epics and features that don’t need to be delivered.

LPM metrics

There should be callouts in your epics that call out value that you expect to see from a metric standpoint. If it is an epic and you want to see performance increase of half a second, for example, these should be called out so you can track them over time.

Also, you want to have some guardrails in place, to make sure you don’t have more work in the pipeline than you can get done. Just like at team level, the more work in process, the less can get done. If things get too complex, you won’t get anything done at all. So, you have to define value that should be trackable, not just at the pie-in-the-sky level.

Validating market impact

Here are a few things to do to validate market impact:

  • Performing Net Promoter Scores ( NPS) to get feedback from customers to determining how does the new functionality work.

  • Defect density to see how many defects come back to determine levels of quality.

  • Running performance tests in background running to determine technical things or architectural things.