Stage gate new product development

New-product development turns uncertain ideas into progressively stronger investment cases. A stage-gate process divides that work into stages separated by decision points. During each stage, a cross-functional team gathers evidence and reduces a defined set of uncertainties. At each gate, authorised decision-makers review the evidence, decide whether the opportunity still deserves investment and allocate the resources required for the next stage.

When to use it

Use a stage-gate process to manage the development and launch of products, services, technologies or substantial product changes when:

• investment and uncertainty increase as development proceeds;
• technical, customer, commercial, operational and regulatory evidence must be integrated;
• several opportunities compete for limited resources;
• explicit go, hold, recycle or stop decisions are needed; or
• senior sponsors need a consistent governance view across an innovation portfolio.

Tailor the process to the risk and novelty of the project. A minor packaging change should not carry the same evidence burden as a new medical device, industrial platform or business model. For highly uncertain digital work, stages can include rapid experiments and agile delivery; governance does not require a slow, linear waterfall.

Origins

Phased project reviews existed before the modern new-product model, including staged investment decisions in large engineering and chemical projects. Robert G. Cooper developed the best-known Stage-Gate® system from empirical research comparing successful and unsuccessful product-development projects. He introduced the model publicly in nineteen eighty-six, and the name Stage-Gate came into use in the late 1980s. Cooper and Scott J. Edgett subsequently refined the approach through several generations, adding portfolio management, differentiated pathways and more adaptive execution.

Stage-Gate® is a registered system associated with Stage-Gate International. “Phase-gate” and “gated development” are also used more generally for related staged-governance processes.

What it is

A conventional process begins with discovery and an initial idea screen, followed by five stages. Each stage contains work; each gate is a governance decision.

Idea screen

Ideas may come from customer research, employees, partners, technology development, competitors, service failures or changes in regulation. The initial screen asks whether the opportunity is strategically relevant, plausibly valuable and worth a small amount of investigation. It should remove obvious mismatches without demanding the evidence that only later work can create.

Stage 1: concept creation

Conduct a preliminary technical and market assessment. Clarify the customer problem, target segment, existing alternatives, strategic fit, rough feasibility and major uncertainties. The output is a concise opportunity statement and a recommendation for deeper investigation.

Stage 2: business case

Define the product or service concept, value proposition, target market, requirements, route to market, development plan and financial case. Test the assumptions behind demand, price, cost, capability and risk. The business case is a living hypothesis set, not a promise disguised as a forecast.

Stage 3: product development

Design and build the product, service and associated operating system. Development includes more than the technical artefact: manufacturing or delivery processes, supply, data, support, compliance, sales enablement and launch preparation must progress together.

Stage 4: test and validation

Verify technical performance, safety, quality, customer acceptance, operational readiness and commercial assumptions. Testing may include prototypes, pilots, controlled market trials, pre-production runs and validation of the economics at realistic scale.

Stage 5: launch and monitor

Move into full production or service delivery, execute the go-to-market plan and monitor adoption, quality, economics and customer outcomes. A post-launch review compares results with the decision assumptions and captures lessons for the portfolio and process.

Developments of the model

A useful gate has four elements:

Defined deliverables: the evidence the project team brings. Explicit criteria: the strategic, customer, technical, financial and risk tests used to judge it. A decision: go, conditional go, hold, recycle for more work, or stop. Resource commitment: people, money and sponsor support for the next stage.

A gate is not a presentation ritual. If every project passes regardless of evidence, the organisation has stages but no gates.

How to use it

One. Define pathways by risk

Create a small number of pathways for different project types: for example, major innovation, platform extension and minor change. State which stages may be combined, which evidence is mandatory and who owns each gate.

Two. Set criteria before evaluating projects

Agree the gate criteria and scoring guidance in advance. Include strategic fit, customer desirability, technical feasibility, commercial viability, execution capability, regulatory or ethical risk and portfolio value. Distinguish criteria that must be passed from factors used to rank attractive projects.

Three. Make evidence proportional

At an early gate, a credible experiment and a range may be enough; later gates require stronger validation. Do not demand false precision from an immature idea, but do not allow confidence and enthusiasm to substitute for evidence.

Four. Use cross-functional teams

Include the capabilities needed to create the whole offer, not just the product. Marketing, design, engineering, operations, finance, sales, service, legal, data and compliance should contribute when their work or risk is material.

Five. Separate advocacy from governance

The team should argue for the opportunity and disclose contrary evidence. Gatekeepers should evaluate and allocate resources. Independent customer research or technical review can reduce attachment to an idea without removing the team’s expertise.

Six. Integrate experiments and agile delivery

Within each stage, run the shortest reliable test of the most consequential assumption. Digital teams may deliver in iterations while the gates govern larger investment and market exposure. A gate should accelerate a well-supported project, not impose avoidable waiting.

Seven. Review the portfolio, not only individual projects

A promising project may still be the wrong investment if it duplicates another initiative, overloads a scarce capability or leaves the portfolio strategically unbalanced. Gate decisions should consider opportunity cost.

Example: protective work gloves

A manufacturer developing a glove for an industrial application must validate both performance and production. Early stages clarify the user, hazards, dexterity requirements and willingness to switch. Development produces prototypes with the intended coating, fabric and construction. Validation then compares the new glove with the incumbent under realistic tasks, recording wear time, grip, comfort, perspiration, durability and safety incidents.

The trial must involve enough representative users and controls to support a decision. Prototype manufacturing should also approximate the eventual process: a hand-made sample can perform well while masking problems that appear at production scale. Before launch, the team confirms worker safety, site permission, manufacturing capability, quality controls, claims, packaging, training and feedback collection.

Some things to think about

Ask whether the evidence burden matches the project’s uncertainty, whether gatekeepers can genuinely stop or redirect work, and whether portfolio constraints are visible when individual projects are reviewed.

Top practical tip

At every gate, identify the assumption that could most seriously invalidate the project and require the next stage to test it directly. This keeps the process focused on learning rather than document production.

Top pitfall

Do not measure process quality by the number of forms completed. Excessive governance slows good projects, while weak gates allow politically protected projects to consume resources long after their case has failed.

Further reading

• Cooper, R.G. (nineteen eighty-six). “Winning at New Products.” Journal of Product Innovation Management, three(4), three hundred and seven–three hundred and eight.
• Cooper, R.G. (nineteen ninety). “Stage-Gate Systems: A New Tool for Managing New Products.” Business Horizons, thirty-three(3), forty-four–fifty-four.
• Cooper, R.G. (twenty nineteen). “The Drivers of Success in New-Product Development.” Industrial Marketing Management, seventy-six, thirty-six–forty-seven.
• Cooper, R.G. and Sommer, A.F. (twenty sixteen). “The Agile–Stage-Gate Hybrid Model.” Journal of Product Innovation Management, thirty-three(five), five hundred and thirteen–five hundred and twenty-six.