Energy flows where intention goes.

Widely attributed to Rhonda Byrne, Australian author, and TV producer

Portfolio Flow

Portfolio Flow describes a state where Lean Portfolio Management provides a continuous flow of new epics to Solution Trains and ARTs to achieve the portfolio’s vision and strategic themes.

The LPM competency aligns strategy and execution by applying Lean and systems thinking approaches to strategy and investment funding, Agile portfolio operations, and governance. This competency has proven to improve business outcomes for SAFe enterprises building the world’s most important solutions. But as with any system, flow can always be improved. Improving the flow of customer value through the portfolio is a key economic driver for the enterprise. That is the subject of this article.

Note: About the Flow Article Series
SAFe is a flow-based system. As such, any interruptions to flow must be identified and addressed systematically to enable continuous value delivery. While flow-based guidance is embedded throughout SAFe, a special collection of eight articles directly addresses impediments to flow. These are Value Stream Management, Principle #6- Make value flow without interruptions, Team Flow, ART Flow, Solution Train Flow, Portfolio Flow, the extended Guidance articles Accelerating Flow with SAFe, and Coaching Flow.


SAFe defines a set of eight flow accelerators (see Principle #6, Make Value Flow without Interruptions) that can address, optimize, and debug issues with achieving continuous value flow. Portfolio flow describes accelerating the flow of the significant initiatives (Epics) needed to accomplish the Portfolio Vision and advance the Enterprise strategy.

#1 Visualize and Limit WIP

Why it matters

Systemic overload caused by too much portfolio work in process (WIP) undermines performance and value delivery. It overloads the ARTs, inhibits responding to change, and results in costly productivity and quality problems, delayed return on investment, employee burnout, and reduced engagement. There is simply no benefit to putting more work into implementation than a portfolio can handle. And nowhere is excess WIP more damaging than deciding which of the most significant strategic initiatives are needed to deliver effective portfolio outcomes.

What to do about it

  • Make all significant Epics visible. Ensure that any projects not transitioned to SAFe and all SAFe epics (above the epic threshold) are tracked in the portfolio Kanban. Otherwise, the total portfolio WIP may be underestimated.
  • Review and adjust the epic threshold to exclude non-epics. Epics are sometimes misused as “important things we want to discuss forever.” Ensure that only significant initiatives above the epic threshold guardrail enter the portfolio Kanban and that there are some reasonable time-bound expectations. Adjust the threshold when needed, which may require including additional criteria.
  • Review and validate portfolio Kanban WIP limits. If there are too many epics in progress, adjust WIP limits. Consider adding classes of service to enable high-priority items to go quickly through the system while lowering the overall WIP.
  • Understand the capacity of each value stream and ART. Collect metrics for each ART’s capacity to prevent more work from going to the teams than they can reasonably analyze and implement.
  • Ignore sunk costs. Stop significant initiatives if they no longer support the strategy or are not providing the intended value.

#2 Address Bottlenecks

Why it matters

Portfolio gridlock can create a bottleneck that hinders the review, analysis, approval, and implementation of the most significant initiatives needed to achieve the portfolio’s vision. The resulting delays threaten the downstream performance of all the Development Value Streams. That creates a choke point that limits portfolio value delivery and imperils strategy.

What to do about it

  • Help ensure LPM has the proper decision-making authority. Verify that the LPM team has sufficient time and authority to participate and make portfolio decisions without escalation.
  • Increase the pool of Epic Owners. A common cause of portfolio bottlenecks is a lack of sufficient Epic Owners. Leverage the knowledge of subject matter experts, Business Owners, Product Managers, Product Owners, and Enterprise Architects to serve as Epic Owners.
  • Understand ART capacity. Portfolio work can’t go anywhere unless there are people to do it. Understand the capacity of each affected ART to take on this new work in a reasonable time frame.
  • Ensure the Lean Business Case is lean. Despite moving to SAFe, many businesses still tend to create traditional business cases that can take too long to develop. Their length and detail, speculative financials, and unproven assumptions may cause questions and delays during analysis.

#3 Minimize Handoffs and Dependencies

Why it matters

Portfolio workflow is critical, but the mechanism of the portfolio kanban is not particularly complex. Typically, there are only a few dozen epics in flight, and they usually require a limited number of stakeholders with the decision authority to advance them. However, managing the work through the portfolio Kanban often requires different skills and people. Some degree of handoffs and dependencies will occur. Bringing the right people into the discussion at the right time facilitates portfolio flow.

What to do about it

  • Properly support the Epic Owner. Epics require review, collaboration, and the engagement of multiple stakeholders who are often busy with their functional roles. Help ensure Epic Owners can reach the people they need promptly.
  • Understand when cross-value stream coordination is required. Some epics will touch multiple value streams. This coordination increases dependencies and handoffs even during the early stages of epic review and analysis. Leverage Epic Owners, Release Train Engineers (RTEs), and Product Management to help decide how this new work can be managed across value streams.
  • Recognize the need to refactor value streams. Much of the new and innovative work that realizes the evolving strategy will arise in the portfolio backlog. During analysis, LPM, with the help of the VMO or LACE, can advise on what value streams may need to be created, adjusted, or eliminated. That, in turn, reduces handoffs and dependencies in the new work.

#4 Get Faster Feedback

Why it matters

Faster Customer feedback, especially in the early stages, is vital to evaluate new initiatives quickly. This feedback ensures epic investments do not result in designing and building solutions that Customers do not want or require a business model change that the enterprise cannot achieve.

What to do about it

  • If epics require changes to the business model, test the assumptions. Epics that impact the business model often come with significant opportunities and big challenges. During epic review and analysis, test business model assumptions with business owners, executives, and Customers helping avoid unrealistic assumptions, false starts, and future sunk costs.
  • Validate viability by engaging with customers early. While solution development resides with the teams, LPM stakeholders need fast feedback about these new initiatives to support decision-making. These decisions should happen long before the MVP by testing early mockups and low-fidelity prototypes. Additional Customer insights should also be captured in the Lean business case.
  • Focus on the leading indicators for MVPs. Verify that epics employ the right leading indicators to measure their progress toward the business outcome hypothesis.

#5 Work in Smaller Batches

Why it matters

Smaller batches go through any system more quickly and with less variability, fostering faster learning. These smaller batches are especially important for the portfolio because epics are significant initiatives that can take a long time to review, analyze and approve. If the batch size is too big, portfolio stakeholders may not have time to evaluate all the work in flight responsibly. This creates delays in decision-making and clogs the flow of value through the portfolio Kanban system.

What to do about it

  • Limit the number of epics reviewed during LPM events. Experiment with how many epics LPM can realistically be considered during the LPM events, such as the Portfolio Sync. Then, adjust the limit based on context and history.
  • Reduce the transaction cost of reviewing and analyzing epics. For example, timebox each epic review to ensure discussions remain clear and concise. This also has the beneficial effect of increasing decentralization and fostering trust.
  • Conduct low-fidelity tests during analysis. Perform research spikes (explainer videos, landing pages, interviews, paper prototypes, fake product sign-ups, micro-surveys, etc.) to understand better the Customer’s problems and the range of solutions.
  • Leverage a common cadence. Applying the same PI cadence across all value streams in the portfolio can help reduce batch sizes, providing a regular rhythm for reviewing portfolio work.
  • Reduce experiment size. Wherever possible, reduce the effort and scope of work needed to get earlier feedback.

#6 Reduce Queue Lengths

Why it matters

Long queues of portfolio work reduce strategy responsiveness and can cause the enterprise to miss critical market windows. This can have a direct effect on the overall competitiveness of the enterprise.

What to do about it

  • Reroute non-portfolio work immediately elsewhere. The portfolio can be a “catchall” for “everything important that needs to be discussed.” Challenge why each item needs to be in the portfolio. Limit the portfolio queue to true epics that require portfolio attention due to their investment or impact.
  • Understand and attend to critical market events and market rhythms. Buying is an act of the Customer, not the producer. Portfolio epics must always be considered in light of the time criticality of the initiative from the Customer’s perspective.
  • Eliminate non-strategic and bad ideas quickly. Since more effort and capacity are needed as an epic moves through the Kanban, LPM should quickly decide which will proceed or be removed.
  • Replace fixed schedules with flexible roadmaps. Inflexible roadmaps create long queues and delays in introducing new critical work. Limit longer-range commitments and replace fixed plans with flexible rolling-wave roadmaps.

#7 Optimize Time in ‘The Zone’

Why it matters

It can be challenging for busy executives to find sufficient time to develop and evolve strategic plans and engage in the ongoing work of LPM. It requires significant intellectual and mental energy, as free as possible from daily business distractions. In other words, it’s hard to find the ‘time in zone’ [1] necessary to collaborate with peers who share responsibility for strategy. What results is a flawed strategy, communicated ineffectively or established and rolled out intermittently versus a continuous evolution.

What to do about it

Leaders, change agents, and coaches need to facilitate a process and experience that creates time in the zone to formulate and execute strategy. Although it’s one of the primary purposes of LPM and its various events, those meetings alone may not provide enough time. Additional focus is required.

  • Ensure sufficient time is allocated to develop the strategy. Consider holding separate portfolio strategy workshops to focus solely on reviewing the portfolio strategy, vision, Lean Budget Guardrails, and business outcome metrics.
  • Hold effective Portfolio events. Ensure the Strategic Portfolio Review, Portfolio Sync, and Participatory Budgeting (see LPM) events are effective and held regularly. Ensure that the data presented are current and provide insights to facilitate decision-making by the right individuals.
  • Recognize when a Portfolio Epic no longer needs LPM focus. LPM and portfolio stakeholders have limited time. Ensure the focus is on epics that require LPM oversight and focused attention. Move epics to ‘done’ when they are rejected or as soon as they are no longer a portfolio concern.
  • Eliminate redundant portfolio governance practices. Epics receive increased scrutiny because of their size, cost, and impact. Replace traditional reporting with SAFe’s Lean-Agile practices that measure progress objectively, including leading indicators and KPIs, to ensure efficient LPM focus.
  • Invest in meeting facilitation to optimize the ‘zone.’ Effective facilitation will address timeboxing, personalities, and misalignment. It can also create a “must-attend event” where people say no to any potential cancellation or lack of attendance. Effective facilitation can also help eliminate or address emotional drama or misalignment amongst key stakeholders.

#8 Remediate Legacy Policies and Practices

Why it matters

Moving to LPM is a significant change for any enterprise. And during the transition, many companies will likely manage some current initiatives by applying their existing traditional governance policies. This means that two different oversight methods will attempt to coexist, resulting in people reverting to the old process for even the new initiatives. This creates an ongoing burden for the enterprise, increases overhead, and slows the flow of value.

What to watch out for

The remedy is to recognize that many historical impediments probably exist and to address them when they occur. Common obstacles include:

  • The portfolio Kanban becomes a central intake for all work, not just new epics that exceed the Lean budget guardrails
  • Funding projects instead of value streams
  • Continuing to forecast and report capital and operating expenses based on outdated practices and manual timesheets
  • Quarterly strategic planning is disconnected from PI Planning and the LPM portfolio review process and includes different stakeholders
  • Making significant customer commitments without engaging with the teams who will do the work or understanding the impact on in-flight or committed work
  • Overly-detailed business cases that require too much upfront investment, burdening and limiting solution design
  • Limiting access or visibility to the portfolio Kanban, causing misalignment, uncertainty about downstream priorities, and redundant work
  • Running traditional program management in parallel or on top of Agile practices. (For example, waterfall phase gates, change control boards, project cost accounting, and outdated status reporting)
  • Incongruent and conflicting policies in finance, HR, etc.
  • Mindsets that defend sunk costs or ongoing commitment to legacy products and services that prevent pivoting to new opportunities

What to do about it

Identify the legacy portfolio activities that should be stopped or replaced. Take appropriate action based on root cause analysis. Ensure reasonable fidelity to the SAFe LPM process.

Measure and Improve flow

SAFe’s Measure and Grow guidance offers portfolios a way to assess and improve their ability to deliver innovative business solutions quickly. It includes six measures specific to flow: distribution, velocity, time, load, efficiency, and predictability. Flow time, load, and distribution are particularly relevant to the portfolio and are described below.

Flow Time

Flow time measures the interval needed for all the steps in a defined workflow to be completed. Portfolio flow time can be estimated from ideation to production. Or for example it can also be helpful to measure portfolio flow time from when an epic is pulled into the ‘review’ state until its hypothesis has been evaluated.

Figure 1 shows an example of portfolio epic flow time. In this example, flow time represents the period it took an epic to go from ‘reviewing’ to ‘done.’ (Note, ‘done’ doesn’t mean the implementation is complete, but it is no longer a portfolio concern.) The scatter plot illustrates that 99% of epics had a flow time of fewer than 261 days, 75% fewer than 168 days, and 50% had a flow time of fewer than 115 days.

Figure 1. An example of portfolio flow time
Figure 1. An example of portfolio flow time

Flow Load

Figure 2. Example Flow load for a portfolio illustrated in a CFD
Figure 2. Example Flow load for a portfolio illustrated in a CFD

Flow load indicates how many items are currently in the system. Keeping a healthy, limited number of active items in portfolio WIP is critical to enabling the fast flow of strategic value. Figure 2 illustrates a Cumulative Flow Diagram (CFD) that depicts the flow load of epics for a specific time frame. The vertical line highlights the current flow load, which excludes the ‘done’ state.

Flow Distribution

Flow distribution measures the amount of each type of work in the system at a specific time. A helpful view of portfolio flow distribution illustrates the trend of money allocation across investment horizons.

Figure 3. An example value stream’s flow distribution for investment horizons over time
Figure 3. An example value stream’s flow distribution for investment horizons over time

SAFe’s flow metrics can highlight improvement opportunities for portfolio work. However, these measurements cannot tell the whole story alone. Qualitative analysis is required to provide context for the flow metrics and further understand the portfolio’s current state.

Learn More

[1] Stillman, Jessica. 3 Tricks to Help You Get in the Zone., September 9, 2015. Retrieved October 12, 2023, from

Last updated: 12 October 2023