Six gatesin practice.
This note walks through a candidate that passed five gates and was declined at gate four. The decline is the mechanism working, not a system failure. A curation framework that never says no is a marketing framework.
The candidate
In late February a candidate reached gate review: a premium spice category, a cooperative supplier with three prior cycles on the platform (all within-base), a proposed 90-day cycle into the UK and EU markets. Call it candidate #2031, the real identifier stays on the internal record.
On paper, the candidate looked strong. The supplier had a clean scorecard. Category demand signals at gate 1 passed comfortably. Supplier vetting at gate 2 was a refresh rather than a full pass, which is normal for a third-cycle supplier. Margin modelling at gate 3 showed a healthy conservative anchor at +9.8%, base at +13.1%, optimistic at +17.4%, well above the platform threshold.
Then gate 4 ran.
Gate 4, competitive scan
Gate 4 is the live-market scan. It looks at current pricing, stock depth and saturation across the candidate resale lanes. It is deliberately separate from the margin model, the model is built on the commerce desk's assumptions before the cycle is priced; the scan is a reality check against what is in market that week.
Three things came out of the scan.
What the scan found
Compression on the price corridor. Two regional distributors had adjusted their list price on comparable grade in the previous six weeks. The adjustments brought the channel-partner gross margin assumption in the candidate's model below the platform's conservative floor, by a meaningful margin, not a rounding difference.
Saturation on the headline channel. The specialty retail lane the candidate was leaning on had absorbed a larger cycle from a different origin two weeks earlier. Stock depth reports indicated the category was well-covered for at least 60 days.
Seasonality risk. A large programme in the destination was scheduled to end during the cycle's resale window. The programme had been supportive of pricing; its absence was expected to soften realised prices further through the back half of the window.
The decision
The committee discussion ran about 40 minutes. Two reviewers argued the cycle could still clear, the conservative case was generous and the supplier's prior cycles suggested operational reliability would compensate for some channel weakness. Three reviewers disagreed: the scan findings pointed toward a real possibility of a below-base outcome on a cycle that didn't need to be taken.
The committee voted to decline. The decision was written up, counter-signed and logged against the candidate record.
What happens after a decline
The candidate record is archived, not deleted. The supplier is informed of the decline and the specific findings, a decline is a curation output, not a supplier judgement. The supplier's existing scorecard is unaffected.
If the same category is proposed again, either by the same supplier or another, the previous decline record is part of the next gate-4 review. If the market has clearly shifted (new programmes, different saturation profile), the candidate can come back. In this case, the commerce desk flagged a potential revisit in Q3 2026 when the programme in question typically cycles back.
Why the platform publishes this
Roughly 40% of candidates get declined at one gate or another. That number is only useful if people know what it resembles concretely. This note is what "gate 4" resembles in practice, a specific set of findings, a specific committee discussion, a specific written output, a specific archive entry.
Curation discipline is not proven by the cycles on the book. It's proven by the ones that never reached it.
Forty percent declined.That's the discipline.
Read the cycles that made it through. The brief carries the same evidence that the committee saw.