Improving Strategic Investment Impact by Limiting Portfolio WIP
by Eric Willeke, SAFe Fellow, SPCT, Elevate.to
Note: This article is part of the Community Contributions series, which provides additional points of view and guidance based on the experiences and opinions of the extended SAFe community of experts.
A few years ago I wrote an advanced topics article (then called a “guidance article”) titled “A Lean Perspective on SAFe Portfolio WIP Limits” that explored the different ways in which SAFe limited work in process at the portfolio level. However, that was SAFe 3.0, and things have changed. In recent versions, especially in 4.6 and 5.0, SAFe provides significantly improved techniques for steering the portfolio through Strategy, Vision, and lean budgeting Guardrails, in addition to introducing the Large Solution (previously Value Stream) level as a tool for managing large-scaled dependencies between ARTs. In addition, the “why” behind limiting WIP has become clearer for me personally, and this has changed the relative importance and preference of using the various WIP-limiting tools.
Before diving in, it’s worth clarifying language around what I mean by WIP limits. In common usage, a “WIP Limit” refers to the number at the top of a column of a kanban board that indicates the maximum number of cards allowed to be in that column at any time. However, in systems design WIP is a larger consideration, and it can’t be relegated to a single column limit in a kanban. There are a number of different thinking tools we can use to manage how much work is in the system, and it is to that broader set of tools that I refer in this article when I talk about “Limiting WIP”
Why do we limit portfolio WIP?
This article primarily addresses limiting WIP at the portfolio level. After all, if it doesn’t happen there, it will overload all the downstream activities, no matter how clever the teams’ kanban systems are. It is a primary consideration in any lean system and SAFe Principle #6 describes it as one of the three keys to creating flow. When we reduce WIP, we see higher productivity, better quality, and more fit-for-purpose outcomes than when we don’t. But what has perhaps failed to be asked often enough in the past is “to what end?” Why are we wanting those things? In practice, there are three top level reasons, depending on what investment horizon you’re investing into. Horizon 3 limits WIP so we can test more hypotheses faster in order to accelerate exploration. It can actually accelerate innovation, because the most important items get attention more quickly. Horizon 2 and the invest segment of Horizon 1 limit WIP so we can create more speed, thus getting to market faster and out-performing the competition. The extract segment of Horizon 1 limits WIP in order to decrease cost and improve the efficiency of investments in a space where we would ideally not need to invest at all. All of these impacts are globally useful, but take different precedence depending on context, and change what follow-on decisions you make as a result of receiving the benefits of limiting WIP.
SAFe is rich in both thinking and practice tools to support all of these goals, especially at the team and ART scopes. However, one of the key mindsets of WIP limiting must be seeded at the portfolio, and failure to apply it via Lean Portfolio Management (LPM) will prevent the entirety of the portfolio from gaining the desired benefits. That key mindset is “focus”, which is a different way of framing the concept of limiting WIP. As the amount of WIP decreases, the organization is naturally more focused on what remains, and the decisions on which investments receive that increased focus becomes more impactful and important. This investment decision is the essence of how LPM drives strategy.
This article addresses two categories of practices for applying these mindsets at the portfolio level. As Figure one demonstrates, the first set of practices helps narrow the set of options under consideration by helping the organization focus on the work that will accelerate the current strategy. The second set of practices then encourages flow of value delivery by applying more traditional WIP management approaches to the Portfolio Kanban.
Creating focus through strategy
The Strategy and Investment Funding section of the LPM competency article describes four mechanisms to steer strategy in the SAFe portfolio. The use of those mechanisms distills to providing the answers to three key questions:
- What is strategically important to our business?
- To what degree will we apply people and resources to those things?
- To what extent will we starve other things?
Collectively, these three questions provide the core of managing investment focus for the entire portfolio. In this usage, managing investment focus is essentially synonymous with limiting strategic WIP. SAFe provides the desired strategic focus through six primary means, each of which provides a certain degree of WIP limiting described below. Once these tools are successfully applied, it becomes much safer to make WIP-limiting decisions elsewhere in the organization because of the alignment and clarity of strategic intent.
Strategic Themes (including OKRs): The effectiveness of strategy deployment is most often seen in its ability to support the removal of other work that could impede it. The Strategic Themes in the SAFe portfolio provide strong, visible anchors to the outcomes and pursuits that are considered most valuable in the current context of the Enterprise. This should provide a means for leaders at every level to push back against new and existing work that isn’t aligned to those goals, and support the refocusing of resources from context core, as described throughout Dealing with Darwin. 
Align around Value: The realignment of ARTs and Solution Trains to focus on specific, value aligned areas provides a very powerful tool for eliminating the distracting work and focusing on the critical things. Component-based groups and functional-pools allocated to projects make it far harder to see when misaligned work is being fed into the system. As a result, aligning around value as described in principle #10 is the second most powerful tool for creating strategic focus.
ART Visions and ART Canvas: While value alignment itself provides a very strong focusing tool, that tool can be amplified by providing a strong anchoring vision for each individual ART, coupled with a well-maintained ART canvas that ensures very strong clarity of that value. In the same way the Strategic Themes tell the portfolio what isn’t core to strategy, the ART vision can inform the release train what is and isn’t valuable at a finely grained level.
Value Stream Funding: While often perceived as interchangeable with aligning around value, the set of funding tools provided when aligned around value add an additional layer of WIP management tools. With Value Stream Funding, LPM leaders are capable of limiting the overall work done against each set of goals, which are already encapsulated in the ART canvas.
Business Owner Engagement (guardrail #4): The fourth guardrail provides an even more focused tool for understanding what is core to the goals of a train and what isn’t. Having the business owners responsible for receiving value from Solutions present in the decision process on a routine basis is a very powerful tool for removing distracting work, especially when the guidance received is constrained by the ART vision and Portfolio Strategic Themes.
Epic selection and approval: The final tool for creating strategic focus is the selection of which epics are approved for implementation. When the ARTs are well designed and operating with flow, the ability of the LPM to exercise restraint in introducing new epics is a powerful tool for limiting how much disruption is created across the portfolio, allowing each individual epic to be more effective in creating the positive disruption that accompanies strategic changes.
Limiting Portfolio Epic WIP with the Portfolio Kanban System
The scope and authority associated with an epic is capable of causing massive disruption to flow and quality due to the volume of other work each epic can displace. Oversupply of epics risks “dueling epics” where the organization receives competing messages about which epics are most important, creating thrash and dependency misalignment across various ARTs. These issues can easily introduce friction that destroys the benefits meant to be provided by epic-based alignment, and often lead to reintroducing project-like behaviors to compensate, which in turn extends lead times, increases costs, and decreases fitness-for-purpose.
Otherwise, the tools for WIP limiting Portfolio Epics are embedded in the Portfolio Kanban, although most of them do not have explicit count-based WIP limits. Below, we will describe each state of the kanban through the lens of how it supports limiting WIP.
Funnel: The funnel limits epic WIP by challenging strategic alignment and “appropriateness” to be an epic. This should start driving an awareness across the organization and help bring the strategic themes to life. While “all ideas are welcome”, not all should move forward, and the funnel should be cleaned up frequently. Don’t let the funnel become a long-term parking lot for bad ideas. Instead, identify items to remove, using the opportunity to have a mentoring and learning conversation with the ideator where appropriate to support organizational learning and individual growth.
Analysis & Review: These two states of the kanban are explicitly WIP-limited because the work of analyzing and creating solution alternatives itself consumes investment and causes distraction. The specific limit permitted represents a decision by the LPM executives of how much money and distraction is appropriate.
Portfolio Backlog: While there is no explicit limit set to the portfolio backlog size, in practice it has an implicit limit based on desired aging and investment flow characteristics. If items in the portfolio are consistently being passed in the final cost of delay sequencing, they are candidates to be pulled out and eliminated or passed back through the analysis flow for reconsideration and rescoping. Remember, the portfolio backlog is not a linear queue, instead, it is continuously reprioritized as newly approved epics arrive.
Implementation (MVP): The number of epics that are currently pursuing experiments to test their strategic hypotheses is limited by the ART and organizational capacity for experimentation, with the risk of over-WIP arising when Horizon 1 work is unintentionally disrupted by the new, shiny, fun experiments. This balance could be managed case by case, or steered through funding guardrail #1: Guiding investments by horizon.
Implementation (Persevere): Work in implementation is fundamentally limited by the implementation ‘pull’ system and the capacity of the ARTs to advance the work associated with those epics. The choice of retaining them in the Portfolio Kanban to maintain executive visibility does not specifically change the overall level of WIP in the system, although it does potentially introduce friction, overhead, and the risk of over-centralizing decisions in a project-like manner if not limited to items of truly strategic nature.
Done: There’s no real value in having limits around the Done state. After all, it’s done from a portfolio perspective. However, it’s a useful mechanism to remind executives that every investment decision into new technology carries a cost well into the future for maintaining and operating that technology.
 Moore, G. A. Dealing with Darwin: how great companies innovate at every phase of their evolution. New York: Portfolio, p xii, 2008.
Last update: 11 June 2020