Sell-through, velocity, and building a buyer case that survives scrutiny.

System 05 of six · Read time about nine minutes

Shelf space is finite and already occupied. Every facing you are given is a facing taken from an item with a known sales rate, and the buyer who takes it has to defend that decision internally. So the demand system is not about enthusiasm. It is about producing evidence, in the unit of measure the buyer works in, that your item will do better than the incumbent.

That unit of measure is almost always units per store per week. If your case is built in any other currency — total revenue, growth rate, social following, review count — it has to be translated before it means anything, and if you do not translate it the buyer will do so unfavourably or not at all.

The meeting is the middle of the evaluation, not the beginning

By the time you are in the room, a competent buyer already knows a great deal. They have looked at your marketplace price against your proposed shelf price. They have checked whether your MAP holds. They have looked at syndicated data if your category has it, and at your reviews if it does not. They have formed a view on whether you have studied their business.

This changes what the meeting is for. It is not a reveal. It is a test of whether the story you tell matches the evidence they have already gathered. Brands that walk in with a narrative contradicted by public data lose credibility in the first ten minutes and never recover it, regardless of how good the rest of the pitch is.

Four questions you will be asked

1. The velocity question. “What turns do you project per door per week?” Arrive with a specific number and the assumption set behind it. Not a range, not “it depends.”
2. The margin question. “What margin can you support?” A number. Not a range, and not “we’re flexible” — flexible reads as unmodelled.
3. The distribution question. “What is your current footprint and why us next?” This tests whether you have a channel strategy or an opportunistic habit.
4. The competitive question. “What is currently in our set?” Generic differentiation fails here. They mean their specific shelf, in their specific stores.

All four are answerable in advance. None of them can be improvised well.

Building a velocity model you can defend

A defensible model is not a precise one. It is one where every assumption is visible and each can be argued with individually. Build it in this order:

• Category rate. What does a comparable item in this set actually do per store per week? Syndicated data if you have it; observed facings and restocking cadence if you do not.
• Your relative position. Are you a better version of the incumbent, a cheaper one, or something the set does not have? Each implies a different multiplier and you should state which you are claiming.
• Distribution and facings. Number of doors, facings per door, and where in the set. A second facing is not a doubling; a bad shelf position can halve the estimate.
• Support. What promotional activity is assumed, and what happens to the number without it. Buyers discount unsupported estimates heavily, and correctly.
• Seasonality. Weekly, not annual. The eight-week window that decides your fate may sit in your worst quarter.

Then state the number, state the range, and say which assumption you are least confident in. Volunteering your own weakest assumption is disarming and it is what an operator does. Hiding it is what a salesperson does, and buyers meet salespeople all day.

Translating DTC and marketplace evidence

Your e-commerce data is genuinely valuable and almost never usable in raw form. What translates:

• Repeat purchase rate — the strongest available proxy for whether an item will re-turn on a shelf rather than sell once to curiosity.
• Geographic concentration — tells the buyer where the item will index, which maps directly onto a regional test.
• Review sentiment on specific attributes — particularly where those attributes are the ones the category is currently failing at.
• Price elasticity you have actually observed — what happened to units when you moved price, which is real evidence and rare.
• Return rate and reasons — a buyer would rather hear this from you than discover it.

What does not translate: total revenue, bestseller rank, follower counts, growth percentages off a small base, and press coverage. None of these predict units per store per week, and leading with them signals that you do not know what does.

The eight-week window

Once you are placed, the clock starts and it runs faster than most founders expect.

Weeks one to four produce thin data. Nobody can distinguish organic demand from new-item curiosity, and the retailer waits. Weeks five to eight are where the signal appears — repeat purchases, basket consistency, velocity trend. Weeks eight to twelve are the decision window: there is now enough data to project, and strong velocity earns support while declining velocity starts an exit process.

The important structural fact is that the elimination decision begins around week eight, while the conversation about it may not reach you for months. That gap is the entire opportunity. Brands that survive are the ones watching their own sell-through weekly and acting inside the window — fixing a price, funding a demo, correcting a shelf position, resolving a stockout — rather than waiting to be told.

Which means you need access to the data. Ask for it during onboarding, understand the cadence and the format, and assign someone to read it every week. A brand that cannot see its own sell-through is managing blind through the only period that matters.

What good looks like

A working demand system produces: a velocity model with visible assumptions and a stated range; a translation of your existing commercial data into door-level terms; a buyer narrative specific to that retailer’s test rather than a generic deck; weekly access to sell-through with a named reader; and a defined set of actions you will take at week four, six, and eight if the number is under plan.

That last one is what separates brands that get a second season from brands that get a discontinuation email.

Next in the series: System 06 — channel roles, conflict guardrails, and holding decision quality as you grow.