Blog
Portfolio returns are set at the gate, not in the pipeline. How complete, traceable evidence transforms the advance, hold, and kill decisions in NPD.
Most R&D organizations don't lack ideas. PDMA's benchmarking of 651 companies across 37 countries found that only about 60% of launched products succeed commercially — a rate that has held flat across 30 years of methodology improvements. Three decades of better ideation, better templates, and better process discipline moved the number almost nowhere.
That stubborn flatness points at something the methodology debates miss. What sets portfolio returns is a smaller set of moments: the advance, hold, and kill calls made at each gate. Get those right and average ideas compound into strong portfolios. Get them wrong and even a brilliant pipeline produces average results — funded projects that should have stopped, and stopped projects that should have run.
Which means the highest-leverage question in innovation management isn't "How do we generate better ideas?" It's "What evidence reaches the room where allocation decisions get made?"
Look at how those decisions get made today. Teams spend the two weeks before a gate gathering documents, chasing status, and building a deck. The review then evaluates the deck — not the underlying evidence. The committee gets thirty minutes with a narrative someone had time to polish, while the knowledge that should shape the decision stays scattered across drives, inboxes, and memory.
What rarely makes it into the room: the rationale behind key technical choices, the prior programs that tried something similar, the customer evidence that conflicts with the forecast, and the assumptions nobody has revisited since kickoff. McKinsey's work on generative AI in portfolio management points at the same gap — the analysis leaders need exists across the organization, but no one can bring it together fast enough to inform the decision.
The result is a quiet double failure. Weak projects advance because their gaps were invisible in slide form. Strong projects stall because their evidence was stronger than their storytelling. Neither failure shows up as a line item, but both compound across a portfolio.
The fix is a different question at the gate. Instead of "Are the required documents complete?", ask "Do we know enough — with validated evidence — to justify the next investment increment?"
Instead of asking "Are the required documents complete?", ask "Do we know enough — with validated evidence — to justify the next investment increment?"
The fix is a different question at the gate. Instead of "Are the required documents complete?", ask "Do we know enough — with validated evidence — to justify the next investment increment?"
That question has four testable parts. Coverage: have the critical questions in each knowledge domain been addressed — market, technical, business, and the rest? Confidence: is each claim a hypothesis, preliminary evidence, or a verified fact? Currency: has the market moved since the evidence was gathered? Integration: does the technical story still match the market story, or have they drifted apart since the last review?
A gate that runs on those four checks surfaces weak bets early, when killing them is cheap, and gives strong bets the conviction — and the resources — they deserve. Knowledge gaps become explicit instead of hiding behind completed deliverables.
This is the operating model Narratize was built for. Every product gets a Product Knowledge Hub where cross-functional teams continuously add knowledge that organizes itself into searchable intelligence. Gate materials assemble from the stage's actual deliverables — PRD, risk assessment, business case, customer findings — with every claim traceable to its source. Reviewers who want to interrogate a number open its evidence instead of scheduling a follow-up.
One industrial equipment manufacturer now orchestrates decision intelligence across a modeled $1.5B of portfolio value on Narratize, running gate reviews where the evidence behind every recommendation is one click away. The decks got faster to build. More importantly, the decisions got better to make — because the committee's attention finally lands on the judgment call instead of the document hunt.
Ideas will keep getting the attention. Evidence deserves the investment.
See what your next gate review looks like when the evidence assembles itself. Schedule a demo.
Schedule a demo and watch your team's expertise become intelligence the whole organization can use.