Portfolio Kanban Flow
Updated 9 May 2026
The Portfolio Kanban system makes initiative flow visible and manageable at the portfolio level. Work moves through four continuous phases, each running in parallel at all times. The portfolio always has initiatives in all phases simultaneously.
Two types of initiatives flow through the same stages: Strategic Initiatives with a multi-year horizon, and Initiatives of approximately six to eighteen months. They share the same stages but differ in tempo, depth of work, and output at each stage.
Strategic Initiatives do not enter delivery as a single unit. As understanding develops through Continuous Discovery — during Investigate and Investment Framing — the Strategic Initiative is continuously broken down into Initiatives, each with its own Lean Business Case and independent operability plan. This breakdown is a learning process, not a planning event. What enters Continuous Delivery is always an Initiative.
Strategic Initiatives
Strategic Initiatives move slowly through the flow. The primary output of Continuous Discovery is the breakdown into Initiatives — each with its own Lean Business Case.
Initiatives
Initiatives (~6–18 months) move through the same stages. The output of Investment Framing is a Lean Business Case ready for Gate 2.
Stages, Gates, and Queues
The Portfolio Kanban uses three distinct mechanisms:
A stage is where active work happens. Each stage has a defined purpose, a responsible owner, and a defined output.
A gate is an explicit go/no-go decision by the portfolio management function. A gate does not open automatically when criteria are met — it requires an active decision. Gates control what enters each queue.
A queue is a prioritized list of initiatives waiting for capacity to become available. Initiatives in a queue are continuously re-prioritized against each other using Portfolio WSJF. Work is pulled from the top of the queue when capacity is available — not pushed in when a decision is made.
Continuous Sensing
The purpose of Continuous Sensing is to make all incoming demand visible and to filter it for strategic relevance before any analytical investment is made.
Intake
Intake is the entry point for all initiative candidates. Its purpose is to receive and register incoming ideas, needs, problems, and opportunities — without evaluating or prioritizing them.
An idea enters Intake as an initiative candidate when it meets two flow policies:
Who may submit: The submitter must have sufficient organizational mandate to sponsor a portfolio-level investment. Intake is not an open suggestion box — initiatives represent investments of significant scale and require a responsible sponsor.
What is required: A problem or opportunity statement of sufficient substance to be assessed at portfolio level. This is a format requirement, not a strategic filter. A raw idea that cannot be described as a problem or opportunity worth considering at portfolio level is not yet an Intake item.
The submitter indicates an initial classification — Strategic Initiative or Initiative — based on their assessment. This classification is a hypothesis, not a decision, and may be revised in later stages.
See: Needs Capture for how ideas are registered and prepared before entering Intake.
Triage
Triage is a rapid relevance assessment. Its purpose is to protect discovery capacity from being consumed by initiatives that lack strategic or tactical grounding.
Who does it: A Triage Group — senior representatives with sufficient domain knowledge and organizational mandate to assess strategic and tactical relevance. The composition may vary between organizations. The Triage Group prepares a recommendation; the portfolio management function makes the formal Gate 1 decision.
What is assessed:
- Is this initiative relevant given current strategic and tactical direction?
- Is it at the right level — a portfolio initiative, not a feature or an operational decision?
- Is the problem or opportunity statement sufficient for Discovery to begin?
- Does the initial classification as Strategic Initiative or Initiative appear reasonable?
Three outcomes: proceed to Gate 1, park for later reconsideration, or stop.
Triage is deliberately rapid. It does not produce an analysis — it produces a recommendation.
Gate 1 — Entry to Discovery Queue
“Is this initiative worth investing discovery capacity in?”
Gate 1 is a relevance filter, not a quality gate. The portfolio management function makes the formal decision based on the Triage Group’s recommendation.
A go decision places the initiative in the Discovery Queue as a candidate for discovery work. It does not guarantee a start date — that is determined by capacity availability and relative priority in the queue.
A no decision is either a stop or a park — both are explicit.
Discovery Queue
The Discovery Queue is the prioritized list of initiatives approved for discovery work but not yet started. Initiatives compete for discovery capacity based on continuous relative prioritization.
Prioritization is cadence-based and comparative — initiatives are assessed against each other, not in isolation. The mechanism for prioritization is defined in the portfolio ways of working.
An initiative entering the Discovery Queue requires an Initiative Owner — the person responsible for leading the initiative through discovery and beyond. The Initiative Owner assembles the right competencies for each stage of the work as needed. This is not a handoff model; the Initiative Owner and the people involved follow the initiative through its lifecycle.
Continuous Discovery
The purpose of Continuous Discovery is to develop sufficient understanding of an initiative to frame it as an investment and reach a go/no-go decision.
Investigate
Investigate is the stage where an initiative is understood deeply enough to be framed as an investment in the next stage. The goal is not to eliminate uncertainty — it is to reduce it sufficiently for a sound investment decision.
Who leads: The Initiative Owner.
Who participates: The people with the domain knowledge and analytical capability that this specific initiative requires. Composition varies per initiative. For complex initiatives — such as replacing a core system — this may involve significant investigation over time across multiple competence areas.
What happens:
- Understand the problem or opportunity: users, customers, business needs
- Explore possible solution directions without committing to one
- Identify risks and significant uncertainties
- Conduct analyses, interviews, or experiments as needed
- Confirm or revise the Strategic Initiative/Initiative classification
Portfolio management conducts continuous oversight. They may trigger a pivot — a change in direction — or a stop decision if evidence warrants it. They do not own the investigative work.
Output: Sufficient understanding to frame the initiative as an investment. Not a business case — that is the next stage.
Investment Framing
Investment Framing structures what has been learned in Investigate into a form that enables a go/no-go investment decision and allows the initiative to be compared with others.
Who leads: The Initiative Owner, with the same team that conducted Investigate.
What happens:
- Articulate the value hypothesis: what do we believe will happen if we invest in this?
- Define expected business outcome and how it will be measured — linked to strategic goals, KPIs, KRAs, KRIs, or OKR Key Results
- Establish rough size and duration
- Define the MVP: the smallest thing that can begin validating the hypothesis
- Assess risk and identify the plan for independent operability
- Produce the Lean Business Case
The Lean Business Case is a deliberately lightweight investment hypothesis — not a project plan. It must be sufficient to support a go/no-go decision and a relative prioritization against other initiatives in the Delivery Queue.
For Strategic Initiatives, the primary output of Investment Framing is the breakdown into Initiatives — each with its own Lean Business Case and independent operability plan.
Output: Lean Business Case, ready for Gate 2.
Gate 2 — Entry to Delivery Queue
“Is this initiative worth investing delivery capacity in?”
Gate 2 is the investment decision. The portfolio management function makes a formal go/no-go based on the Lean Business Case.
A go decision places the initiative in the Delivery Queue as an approved investment, ranked by Portfolio WSJF against other approved initiatives. It does not guarantee a start date — that is determined by capacity availability and relative priority in the queue.
Gate 2 is independent of current delivery capacity. It is a quality and strategy decision, not a scheduling decision.
Delivery Queue
The Delivery Queue is the prioritized list of approved initiatives waiting for delivery capacity to become available. Initiatives are ranked continuously using Portfolio WSJF and re-prioritized on cadence.
Work is pulled from the top of the Delivery Queue when delivery capacity becomes available in the DVS. The highest-ranked initiative that the DVS can start is pulled first.
The Delivery Queue is the primary tool for managing trade-offs between competing investments. An initiative that consistently sits at the top of the queue but cannot start is a signal that the organization needs to address a capability or capacity constraint — not a reason to start lower-ranked work instead.
Continuous Delivery
Implementing
Implementing is the stage where the DVS delivers the solution. Active delivery work — features, stories, and technical execution — happens at DVS and team level, not at portfolio level.
The portfolio level has three responsibilities during Implementing:
Continuous hypothesis validation: The Initiative Owner tracks delivery against the LBC hypothesis. Is the solution being built consistent with what was hypothesized?
Pivot or persevere decisions: When sufficient evidence is available — from MVP outcomes, changed conditions, or emerging learning — the portfolio management function and Initiative Owner decide whether to continue, change direction, or stop.
Oversight, not management: Portfolio management maintains continuous visibility into active initiatives without managing the delivery work itself. DVS and team-level decisions remain at DVS and team level.
An initiative may be stopped during Implementing if the hypothesis is falsified or conditions change materially. This is not a failure of the process — it is the process working.
Done
Done means delivery is complete and the solution is in production and available to users. Active portfolio governance of the delivery effort concludes.
Done is a transition, not an endpoint. It does not mean value has been confirmed — it means the investment has been made and the conditions for outcome assessment now exist.
Continuous Learning
Outcome & Learning
The purpose of Outcome & Learning is to assess whether the delivered initiative produced the expected outcomes and to feed that learning back into the system.
An initiative enters Outcome & Learning when delivery is complete. It exits when the outcome assessment is concluded — typically after sufficient time has passed for meaningful evidence to emerge, often one full OKR cycle.
What happens:
- Observe outcome evidence against the LBC hypothesis
- Assess movement on the strategic measures the initiative targeted — KPIs, KRAs, KRIs, OKR Key Results
- Interpret what the evidence tells us: confirmed, disconfirmed, or inconclusive
- Feed learning forward: how does this change future investment decisions, strategic priorities, and initiative hypotheses?
Who is involved: The Initiative Owner presents the evidence. The portfolio management function interprets it together with relevant stakeholders. The Strategic Portfolio Review is the primary forum for outcome assessment at portfolio level.
The learning from Outcome & Learning feeds back into Continuous Sensing and Continuous Discovery — informing what the organization looks for, what hypotheses it forms, and how it prioritizes. A system that does not close this loop does not improve its investment accuracy over time.
Related Metrics
Flow time, queue age, WIP, and predictability metrics for each stage are defined in Portfolio Flow Metrics.
Sources
- Reinertsen — Principles of Product Development Flow (2009). Queue theory, WIP management, and the economic basis for prioritization.
- Anderson — Kanban (2010). Pull systems, flow policies, and class of service.
- Humble & Farley — Continuous Delivery (2010). Continuous validation and feedback loops in delivery systems.
- Gothelf & Seiden — Lean UX (2013). Hypothesis-driven investment thinking.
- Cagan — Inspired (2017). Discovery as a continuous organizational capability.