| |
|
PROCESS NOTES · ISSUE #005 · SCALE-UP PLAYBOOK · 10 MIN READ
Stage Gate in Food Development:
How Ideas Get Made — or Killed
A gate-by-gate field guide for R&D and process engineers who build food products
|
| By Galvin Eyong · Food System Expert |
R&D
SCALE-UP
|
|
|
Once a food project gains momentum, nobody stops it. That’s the real problem.
People have been briefed. Suppliers quoted. A launch date dropped into the commercial calendar. The project keeps moving — not because it should, but because stopping it feels like failure. Stage Gate creates structured decision points where leadership actively decides: continue, correct course, or stop. Done well it kills the 14-month projects that should have ended in week six. Done badly it’s a rubber stamp — the decision is already made before anyone walks into the room.
This issue breaks down the full framework gate by gate: what each stage demands, who owns what, the failure modes that bite after the gate says go, and how teams compress timelines without breaking the process.
|
Stages do the work. Gates make the call.
|
|
|
Stages are defined work phases where the cross-functional team executes. Gates are structured decision points where a review panel checks deliverables and makes one call from three options:
|
GO
Proceed to next stage
|
RECYCLE
Fix gaps, return to gate
|
KILL
Stop the project
|
The structure forces three things food teams routinely skip: documented deliverables before each gate, cross-functional sign-off from every function with skin in the game, and an honest conversation about whether this project is still worth the resource.
|
|
The Stage Gate Funnel — Food NPD
|
STAGE 1 · Concept Development
|
|
GATE 1 — CONCEPT
✓ GO · ↻ RECYCLE · ✗ KILL
|
|
|
GATE 2 — FEASIBILITY
✓ GO · ↻ RECYCLE · ✗ KILL
|
|
|
GATE 3 — DEVELOPMENT
✓ GO · ↻ RECYCLE · ✗ KILL
|
|
|
STAGE 4 · Validation & Scale-Up
|
|
|
GATE 4 — VALIDATION
✓ GO · ↻ RECYCLE · ✗ KILL
|
|
|
GATE 5 — LAUNCH
✓ GO LIVE · ↻ RECYCLE · ▮ HOLD
|
|
Typical floor-to-shelf timeline: ~14 months · Stage 6 (post-launch review) at 12 months
|
|
What actually happens at each stage
|
|
|
The gate is only as strong as the work done in the stage before it.
|
STAGE 1 · Concept Development
|
Proving the idea deserves more than enthusiasm
Marketing leads. The team develops the proposition, defines the consumer, writes the marketing mix, and conducts concept testing — 75+ in a quantitative test is the common threshold to proceed. R&D, Quality, and Operations run a first-pass scan: Is it technically achievable? Any novel ingredients or regulatory clearance needed? Does it fit the manufacturing envelope? Commercial Finance sketches top-line financials.
|
Gate 1 checks:
Consumer score above threshold • No obvious regulatory blockers • Commercial case viable • Stage 2 resource committed
|
|
The stage where projects get real — or get killed
The PM stands up the steering team, runs the cross-functional kick-off, and writes the project brief with R&D and Marketing. R&D defines the reference standard product. Procurement screens co-packers and checks MOQs and lead times. The most critical decision is the sourcing and process approach: which co-packer, which manufacturing method, and how shelf stability is achieved. These choices lock the COGS, lead time, and technical risk profile. Commercial Finance builds the feasibility P&L. Risk register opened.
|
Gate 2 checks:
Technical approach confirmed • Sourcing decision made • P&L within margins • Risk register complete • Timeline agreed
|
|
Making it real at bench and pilot scale
R&D leads. Pilot batches run iteratively toward a recipe that simultaneously meets the sensory brief, shelf life target, nutritional declaration, cost target, and claim set. Consumer testing on the prototype confirms the concept score holds. Regulatory completes the claim file. Allergen assessment updated. Procurement confirms final pricing and lead times. Operations drafts the manufacturing plan. Volume forecasts enter demand planning. Phase-in/phase-out plans drafted. Artwork briefed in parallel — waiting until after Gate 3 is one of the most common timeline killers.
|
Gate 3 checks:
Recipe locked at final inclusion levels • Consumer prototype score confirmed • COGS within business case • Manufacturing plan signed • Launch date committed
|
|
|
STAGE 4 · Validation & Scale-Up
|
Proving it works at the scale that matters
Three full-scale production trials is the industry minimum. Each generates data on yield, waste, cycle time, and process capability (Cpk for critical parameters). Microbiological validation confirms food safety controls work at commercial scale — not just pilot. Shelf life testing runs in parallel: accelerated data provides early direction, but real-time samples are mandatory before any declared date is validated. Transit trials simulate distribution stress. Final raw material specs signed. Artwork sign-off: QA checks label vs spec, Regulatory confirms claims, Brand signs design — in that order, independently.
|
Gate 4 checks:
Min. 3 successful production trials • Micro validation complete • Shelf life data meets target • Final COGS confirmed • All sign-offs in place
|
|
From warehouse to shelf — and the review that follows
Materials ordered. The first commercial trial run is a validation run — not automatically the first production batch. Technical approval from R&D, QA, and the site is required before release for sale. Full production follows. Stock ships to the DC on the sell-in date. Product reaches shelf. The post-project evaluation (PPE) runs within 90 days. The post-launch business review at 12 months compares actual performance against the P&L committed at Gate 2. Both are mandatory, not optional.
|
Gate 5 checks:
First commercial trial approved • All logistics confirmed • Retailer PO received • Complaint handling updated for new product
|
|
|
A sparkling adaptogen drink: gate by gate
|
|
|
A beverage brand launches a 4-SKU sparkling adaptogen drink range. Format: 250ml can. Markets: UK and Netherlands. Co-pack manufactured. Year 1 NSV target: €1.8M at 35% margin.
| Gate |
Key decision & outcome |
| Gate 1 — Concept |
Concept scores 87/100 UK, 82 Netherlands. One adaptogen flagged as borderline Novel Food — parked. Three SKUs proceed. GO. |
| Gate 2 — Feasibility |
Tunnel pasteurisation selected. Co-packer selected on line capability and unit economics. P&L within margin targets. 12-month window committed. GO. |
| Gate 3 — Development |
Can liner compatibility testing identifies migration risk — alternative liner selected. Recipe locked. COGS confirmed. Artwork briefed in parallel. GO. |
| Gate 4 — Validation |
Three production trials complete. Micro validation confirms 6-log reduction. Accelerated shelf life supports 12-month target. Artwork: QA, Regulatory, Brand sign-off in sequence. GO. |
| Gate 5 — Launch |
First commercial trial approved. Full production run. Stock in DCs in both markets within 2 weeks. LIVE. |
|
| 04 |
Cross-Functional Ownership |
Who does what, and when
|
|
|
The most common Stage Gate failure is unclear ownership. One team thinks another is handling a deliverable. Clarify this at every kick-off and revisit at each gate.
| Function |
Primary responsibilities |
| Project Manager |
Kick-off, team, risks, timeline, budget, gate submissions, post-project evaluation |
| Marketing |
Proposition, consumer testing, 6Ps, brief, artwork, launch comms, post-launch review |
| R&D |
Technical brief, reference standard, formulation, trials, shelf life, spec sign-off |
| Food Safety & Quality |
HACCP, regulatory compliance, micro validation, allergen, artwork QA sign-off |
| Operations |
Sourcing decision, MOQs, capacity, manufacturing plan, trial execution |
| Commercial Finance |
P&L at every gate, COGS validation, post-launch tracking |
|
The gate said GO. Six months later it bit back.
|
|
|
Five common ways food NPD projects pass their gates and still fail in market.
|
⚠ Landmine 01 — Shelf life declared on accelerated data alone
Gate 3 approved 12 months from modelling. Real-time samples at month 8 show degradation below claim tolerance. You are managing label compliance risk on product already distributed.
|
|
⚠ Landmine 02 — Validated on the pilot, scaled on the commercial line
Different residence times, pump capacities, or heat exchanger configurations mean your thermal validation does not transfer to the commercial line.
|
|
⚠ Landmine 03 — Artwork approved by Marketing, not QA
The allergen statement on the approved artwork doesn’t match the spec. QA wasn’t in the sign-off chain. First production run on hold pending a full re-print.
|
|
⚠ Landmine 04 — Regulatory status assumed, not confirmed
A botanical ingredient requires Novel Food authorisation, not just an assumed positive-list status. Launch delayed 18–36 months. Regulatory was a Gate 2 checkbox, not a critical-path item.
|
|
⚠ Landmine 05 — The trial run became the production run
Marketing pushed to keep running after the trial without re-triggering formal sign-off. Complaints arrived three weeks later — a sealing defect that should have stopped shipment at the approval stage.
|
|
Match the process to the project
|
|
|
Not every project deserves all five gates. Match process intensity to risk and complexity:
|
New product launch. Novel ingredients, new category or format, new manufacturing process. All 5 gates mandatory. No shortcuts.
|
|
| REDUCED · Medium Complexity |
|
Line extension. New flavour or pack count on an existing recipe platform. Sequential gates can be merged with PMO sign-off.
|
|
| FAST TRACK · Low Complexity |
|
Artwork change, sales extension, simple rework. Feasibility gate only — all others bypassed.
|
|
|
| 07 |
Speed Without Sacrifice |
7 ways to compress your timeline
|
|
|
Stage Gate isn’t the enemy of speed — bad Stage Gate is. Every compression below requires explicit stakeholder sign-off to be defensible.
| Acceleration tactic |
Sign-off required |
| Engage production site during Feasibility, not Development |
None. Saves 4–6 weeks. |
| Merge two sequential gate reviews |
PMO sign-off + gatekeeper agreement. |
| Order long-lead materials before gate approval |
Gatekeeper approval + write-off risk in writing. |
| Brief artwork before final recipe lock |
Regulatory alignment on must-have claims first. |
| Launch with conservative shelf life, extend post-launch |
R&D and QA sign-off. Real-time data builds while product sells. |
| Run first commercial trial concurrent with first production |
Full team risk review + hold-and-release procedure mandatory. |
| Split projects with divergent timelines into separate workstreams |
Steering committee approval. |
|
|
Process Notes
Plain-language process engineering for the food production floor. Published weekly via Beehiiv.
Forwarded this? Subscribe free. Got a topic request? Hit reply.
© Process Notes 2026 · Unsubscribe
|
|
|
