Reading amargin model.

What a margin model is and isn't

The margin model on a Treviya brief is not a forecast, a guarantee or a promise. It is a written statement of the commerce desk's assumptions before the cycle is approved, what needs to be true for the cycle to clear at each of three outcome levels.

The model is built during gate 3 (margin modelling). It gets challenged at gate 4 (competitive scan) against what is in market that week. It gets signed off at gate 6 (committee approval) with the written assumptions attached. By the time it sits on the public brief, it has been argued over.

The three scenarios

Conservative case. Built on downbeat assumptions, slow channel uptake, realised prices in the lower quartile of the competitive scan, handling costs at the upper end of the platform's recent range. If the cycle clears the conservative case, the platform has reasonable confidence the cycle will not embarrass anyone.

Base case. The commerce desk's best estimate. Typical channel uptake for the category, realised prices in the middle of the recent range, handling costs at the platform average. The base case is what the desk expects if nothing unusual happens.

Optimistic case. Upbeat assumptions, faster uptake, realised prices in the upper quartile, handling costs at the lower end. Useful for understanding the upside shape. Not useful for sizing commitments.

Why the conservative anchor matters

The platform's committee rule is simple: the conservative case must clear the platform threshold. If the conservative case doesn't clear the threshold, the cycle is declined at gate 3, before the public brief is even drafted. The threshold is not published because it changes with platform conditions, but the principle is fixed.

That means every cycle on the public book has a conservative case that the platform has committed to clearing. Reading the conservative case is reading the lowest outcome level the platform is willing to ship.

Readers who anchor on the base case can still size correctly, the base case is where the desk expects the cycle to land. But they are anchoring on an expectation rather than a commitment. That's a fine choice as long as it's a conscious one.

Common misreadings

Averaging the three scenarios. Doesn't work. The scenarios are not sampled from a distribution; they're three different sets of assumptions. The average is meaningless.

Anchoring on optimistic. Always wrong. Optimistic is there to show upside shape, not to size commitment. A reader who anchors on optimistic is setting themselves up for disappointment on any cycle that lands at or below base.

Reading the numbers and ignoring the assumptions. The assumptions sit next to each scenario on every brief. They are the point. A cycle where the base case assumes "recent category uptake" is a different cycle from one where the base case assumes "softening demand with offsetting price compression." Same base-case number, different risk profile.

What cycles do

Across the platform's closed cycles, the distribution is roughly as follows, this is empirical and will shift as the platform matures:

• Above base, about a quarter of cycles. Factors: faster-than-modelled channel uptake, supplier outperformance or favourable pricing shift during the window.
• Within base, roughly half of cycles. The desk's expectation proves broadly correct.
• Below base, about a quarter of cycles. Factors: channel softening, incumbent price moves, labelling or handling overruns or a combination.

Cycles that land below base case still settle. Cycles that land below the conservative case happen less often, but they happen. When they do, the ledger carries the explanation line-by-line. Cycle #39 (Koroneiki olive oil to DACH) is the most recent example and is covered in its own field note.

Three-word summary

Read the conservative case. That's the number the platform has committed to clearing before the cycle reaches the book. Everything else on the model is context.