The Phase Gate Process: Gate Criteria, the Four-Way Decision, and the Agile Hybrid

Most phase gate reviews run on gut feel. Teams show up to defend projects rather than have them evaluated against a shared standard. Without that standard, nobody really knows what “ready” means, so some projects stall longer than they should while others advance when they shouldn’t.

The phase gate process is the governance layer between project stages—not the stages themselves, but the criteria a project must clear to earn resources for the next one. By the end of this guide, you’ll have a per-gate question set, a working split between must-meet and should-meet criteria so the two never get conflated, and a clear picture of what belongs in a gate packet when the phase behind it ran as Agile sprints.

Höjdpunkter

  • Must-meet criteria eliminate fatal flaws, while should-meet scorecards rank surviving concepts by commercial value and technical feasibility across the portfolio
  • Gate reviews drive forward-looking resource choices rather than backward-looking project audits, funding future value instead of past sunk costs
  • Review boards must choose from four clear verdicts at each gate—Go, Kill, Hold, or Recycle—to stop stalled work from draining company capacity
  • Agile delivery fits inside structured phase boundaries, allowing teams to test hypotheses in sprints while leaders govern business checkpoints
solution demo

Planview New Product Development Solution Demo

Deliver products faster. Close the gap between roadmap and reality.

Watch the solution demo • Planview New Product Development Solution Demo

What Is the Phase Gate Process?

The phase gate process is a governance framework that divides a project into distinct stages. It controls risk and capital commitment from conception to launch. By dividing a project into phases, leadership ensures teams validate key assumptions before spending additional funds.

A phase gate typically evaluates readiness and business value at the end of each phase. It doesn’t judge how well the team executed the past phase. Instead, it determines whether the project earns continued investment for the next phase.

In modern product development, this framework is widely known as the Stage-Gate® process, a registered methodology created by Dr. Robert G. Cooper, which is used by 54% of physical product firms for NPD (Source: Best practices in new product development and innovation: Results from PDMA’s 2021 global survey). While people often use the terms interchangeably, the generic industry term is phase gate. In this guide, we use the phase gate methodology to describe this universal governance structure.

Phase names vary by project type and industry standards. For instance, Dr. Cooper’s Stage-Gate® process defines six stages with one gate at the end of each of the first five:

  1. Discovery
  2. Omfattning
  3. Affärsanalys
  4. Development
  5. Testing and Validation
  6. Lansering
Stage-Gate® Governance Framework • Dr. Robert G. Cooper
Phase Gate Governance Model planview.com

The Six Stages & Five Gates

Life Cycle Flow Sequential Stage Execution Controlled by Formal Decision Gates
Forward-Looking Resource Commitments
Stage 1

Discovery

Idea Generation
Gate 1
Idea Screen
Stage 2

Omfattning

Quick Assessment
Gate 2
Second Screen
Stage 3

Business
Case

Detailed Plan
Gate 3
Go to Development
Stage 4

Development

Build & Design
Gate 4
Go to Testing
Stage 5

Testing &
Validation

Trials & Pilots
Gate 5
Go to Launch
Stage 6

Lansering

Commercialization
Gate 1
Idea Screen
Gate 2
Second Screen
Gate 3
Go to Development
Gate 4
Go to Testing
Gate 5
Go to Launch

These are different from the five process groups defined by the Project Management Institute (PMI). They’re also different from Planview’s nine-stage model.

If you’re looking for the full stage-by-stage walkthrough of the product development process, read our article on the product development cycle.

The primary benefits of phase gate governance come from forward-looking resource allocation. An effective gate acts as an investment checkpoint. It’s not an administrative sign-off. It’s a business process that assesses whether current market conditions, technical proof, and projected margins justify committing your next tranche of capital.

Although this portfolio governance framework works well in modern settings with Agile teams. Planview hosted Dr. Robert Cooper for a webcast on combining idea-to-launch governance with Agile delivery methods to create a more dynamic, flexible way to get modern products to market. More on that below.

What Criteria Determine Whether a Project Passes a Gate in the Phase Gate Process?

Every gate in the phase gate process applies two distinct evaluation tools:

  1. Must-meet criteria
  2. Should-meet criteria

Conflating these two tools makes the gate review process feel arbitrary. When leaders mix critical flaws with strategic preferences, weak projects advance on executive enthusiasm alone.

Must-Meet Criteria Defined

Must-meet criteria form a pass-or-fail checklist. They act as knockout filters. A project can’t move to the next phase without clearing every item.

These criteria focus on existential risks:

  • Regulatory, safety, and environmental compliance requirements
  • Clear alignment with an approved strategic portfolio category
  • Intellectual property ownership and legal clearance
  • Technical feasibility thresholds

Must-meet criteria get stricter as gates advance. At Gate 1, reviewers judge technical feasibility qualitatively. By Gates 3 and 4, it’s a hard pass-or-fail test.

A project that fails a must-meet item stops immediately. Reviewers should check these binary items early. This practice keeps review teams focused on non-negotiable hurdles before debating scores.

Should-Meet Criteria Defined

Should-meet criteria provide a scored comparison against other initiatives competing for the same resources. They evaluate the commercial strength of projects that pass the knockout stage. Reviewers use weighted scorecards rather than simple checklists.

These scorecards rate multiple business dimensions:

  • Technical feasibility and execution complexity
  • Market attractiveness and target segment size
  • Financial return and margin potential
  • Competitive advantage in the market
  • Strategic fit and portfolio alignment

Dr. Robert G. Cooper and Stage-Gate created two key tools for this step: the Value-Based Scorecard (VBS) and Expected Commercial Value (ECV).

The Value-Based Scorecard weights strategic and financial factors into a single score. Expected Commercial Value calculates the risk-adjusted financial return of an initiative based on:

  • Technical project success rates
  • Remaining development costs
  • Commercial probabilities

Financial inputs often include the Net Present Value (NPV) and Internal Rate of Return (IRR). NPV discounts projected cash flows, while IRR measures capital efficiency.

Clearing must-meet criteria doesn’t guarantee funding. If a project scores below competing initiatives and resources are tight, leaders must defer the project.

To learn how to calculate these values, read our guide on how to quantify and score new product development projects.

Key Questions for Every Gate in New Product Development

Each gate addresses specific risks across the project life cycle. As work advances through each project phase, questions shift from strategic alignment to operational execution.

The table below outlines the core readiness and business-value questions for each gate, using Cooper’s original six-stage model:

Gate Readiness question Business-value question
Gate 1 (Discovery / Scoping) Is the concept technically plausible, and does it have a named owner? Does this concept align with an approved strategic category?
Gate 2 (Scoping / Business case) Is there a documented, drafted business case with well-defined target customer profiles? Does the estimated target market size clear our minimum threshold?
Gate 3 (Business case / Development) Did the team validate the business case with real customer feedback and feasibility data? Are this project’s VBS and ECV scores high enough to compete against other projects awaiting the same resources?
Gate 4 (Development / Testing and validation) Did the working prototype pass internal acceptance benchmarks and quality standards, and is it ready for independent testing? Were there any changes during the development phase that might affect the original financial case enough to warrant a rescore?
Gate 5 (Testing and validation / Launch) Did the product pass field tests, customer trials, regulatory compliance standards, and other criteria set at Gate 3? Is the go-to-market plan fully funded and staffed across sales, marketing, and support teams, not just drafted?

What Are the Four Possible Gate Decisions?

Depending on the answer to each question above, a formal phase gate review process produces one of four decisions:

  1. Go
  2. Kill
  3. Hold
  4. Recycle
Phase Gate Governance Model planview.com

The Four Possible Gate Decisions

Current Stage Execution

Fas 1

Work in Progress

Gate (Phase 1 / Phase 2)

Governance Decision
Go

Move Forward to Phase 2

Clears all must-meet criteria & scores high on should-meet

Kill

End Project in Current Form

Fails a must-meet criterion or shows poor commercial value

Hold

Defer Until Resources Open Up

Meets all criteria, but business lacks immediate capacity

Recycle

Return to Phase 1 to Fix Deliverable

Missed specific deliverable, but business case remains strong

Next Stage Commitment

Fas 2

Resources Released

The table below summarizes each decision in the review process:

Decision When to use it What it means
Go When the project clears all must-meet criteria and scores high on should-meet criteria The project can move forward to the next phase
Kill When the project fails a must-meet criterion or shows poor commercial value The project ends in its current form and can’t continue to the next phase
Hold The project meets all criteria, but the business lacks the resources to fund it right now Defer until later without requiring project changes until resource capacity opens up
Recycle The project missed a specific deliverable, but the underlying business case remains strong Return the project to the current stage to fix the deliverable, then re-enter the gate

Go

A Go decision commits budget and personnel to the next phase of the project. The initiative clears every must-meet item, and its should-meet score ranks high across the active portfolio.

For example, an industrial automation concept clears Gate 3:

  • The project team demonstrated technical viability
  • Customer interviews confirmed demand
  • The ECV ranks in the top 30%

Decision: The board issues an unconditional Go, releasing full funding and headcount to move the project into development.

A Go can also be conditional. The project advances, but it must deliver a specific output, such as completed pilot data, before full development resources are released.

Kill

A Kill decision ends a project in its current form, for one of two reasons. First, the project fails a must-meet criterion, a critical flaw that stops it regardless of commercial merit. Second, it clears every must-meet filter but scores too low on should-meet criteria to justify further investment.

For example, suppose a connected device reaches Gate 4:

  • Independent stress tests reveal a major safety issue
  • Fixing the issue exceeds the approved budget
  • The revised Net Present Value turns negative

Decision: The board kills the initiative because it fails the must-meet regulatory criterion. They reassign project team members to higher-value work.

Hold

A Hold decision pauses work due mainly to resource constraints. The project meets all criteria, but the enterprise lacks the staff or funding to start work today.

Let’s say a team presents a compliance reporting tool at Gate 2:

  • The technical architecture is complete
  • The business case is sound
  • Customer interest is high

The decision would normally be a Go. However, two higher-priority initiatives have already booked all engineering capacity for the quarter.

Decision: The committee places the project on Hold, unchanged, until the next review or until enough engineering capacity frees up to take on development.

Recycle

A Recycle decision returns a project to its current phase or stage to fix a specific deliverable gap. Unlike a Hold, which depends on capacity, a Recycle addresses missing or flawed project data.

For example, an industrial sensor team reaches Gate 3 with a complete business case built on internal assumptions:

  • The technology has cleared feasibility testing
  • Target segments are defined and sized
  • The pricing model has never been tested with a real buyer

Decision: Reviewers Recycle the project back into Business Case work to pressure-test pricing with actual prospects. Nothing else about the project changes, and no resourcing constraint is in play — the gate names one deliverable, and the project returns to Gate 3 once that gap is closed.

How Does the Agile Hybrid Change What a Gate Review Sees?

Agile delivery doesn’t replace gates.

In a webinar, Dr. Robert Cooper and Planview’s Carrie Nauyalis explored applications of a hybrid Agile-Stage-Gate to improve productivity.

In a hybrid model, teams work in iterative Agile sprints inside each phase. Those phases still end at formal gates. Each gate applies the same must-meet and should-meet criteria as we saw before.

The primary shift occurs in the gate packet’s content. Traditional gates rely on static requirement documents and speculative milestone charts. An Agile gate packet contains different artifacts, including:

  • An empirical sprint synthesis detailing validated or disproved user hypotheses, replacing speculative specifications with documented test results
  • Capacity requests based on demonstrated sprint velocity and historical cycle times rather than arbitrary upfront estimates
  • Demonstrations of working software, physical prototypes, or tested components that reviewers can inspect directly
  • Updated, dynamic risk logs, rather than a single end-of-phase risk assessment
  • The updated business case reflecting anything learned mid-phase

This phased review process helps organizations balance product and process demands during rapid project execution. It establishes a new gate review procedure that tracks real project progress without stalling sprint velocity.

Timing Gate Reviews in Agile Hybrid

Enterprises often align gate reviews with cadence models. For example, some organizations using the Scaled Agile Framework (SAFe) schedule gates at program-increment boundaries. While not a requirement, this is one way teams synchronize sprint cadence with gate timing.

This schedule aligns executive funding reviews with team planning cycles without breaking sprint momentum.

Why Gate Governance Needs Portfolio-Level Visibility

Evaluating gates one project at a time creates blind spots across your business. When leaders review a single gate in isolation, they can’t tell if approving it will pull critical specialists away from other top priorities.

To make the best decisions, PMO leaders, R&D portfolio leaders, product leaders, and program managers need to see every project’s gate status across the entire portfolio at once. This governance also requires portfolio visibility across the entire project lifecycle.

This means that, for all active programs, leaders need a central view of:

  • Resource demands
  • Schedule health
  • Projektstatus

This visibility helps reviewers decide whether to give the Go and commit capital to move a project forward, or to Hold low-priority efforts.

Modern portfolio management platforms like new product development deliver this portfolio visibility. They provide a single view of each active project’s current gate status, so you can decide to Hold at one gate while seeing what else is competing for the same resources, instead of reviewing each gate deck in isolation.

By connecting day-to-day team execution with executive gate reviews, these platforms help organizations remove silos, balance workloads, and protect their innovation process.

Making Gate Reviews a Decision, Not a Ritual

A gate is a resource decision built on must-meet and should-meet criteria. It’s an investment decision, not a routine status update. Gate reviews connect funding and staff to proven business opportunities through one of four possible decisions.

Go and Kill decisions are easy enough to understand. However, Hold and Recycle decisions solve distinct problems. The former solves a resourcing problem, while the latter solves a specific deliverable gap. Mixing them up costs reviewers credibility.

Agile sprints change what evidence reaches the gate. They don’t remove the gate.

To make your gate reviews decisive, follow three rules:

  1. Separate must-meet from should-meet criteria
  2. Enforce the four-way decision framework at every gate (Go, Kill, Hold, or Recycle)
  3. Let teams run sprints inside phases and use Agile evidence at gates

These practices give teams a defensible basis for every funding decision.

After establishing your gate criteria, the next step is gaining real-time visibility across all active initiatives. Learn how new product development provides portfolio-level gate visibility to help PMOs and portfolio leaders balance resource capacity and accelerate time to market.