CLCTN · STRATEGY NOTE · THE VINTED MODEL, APPLIED · SEPTEMBER 2026
Vinted monetises the flux. We monetise the stock.
Thinking out loud about the Vinted playbook and which parts of it belong in CLCTN. The useful finding is not the growth story — it is that their entire engine is built on one asymmetry they never resolved: people look at the stock in their closet and never at the speed they refill it. That unresolved half is the market we are already standing in.
Written as a note, not a plan. Section 03 is the part to argue with; section 06 is the part to decide.
00 · THE THESIS IN ONE PARAGRAPH
Vinted got rich by making it free to let go of an object and charging a few cents on the way back in. CLCTN can invert every term: it is free to keep an object, and what costs money is everything you want to do with a collection you have decided to keep — know what it is worth, prove you own it, insure it, lend it, and eventually sell it on your terms rather than into a feed of eighty-three million Zara listings.
SAME STRUCTURAL BET (one engine, N object types) · OPPOSITE SIDE OF THE LOOP
01 · THE MECHANISM, STRIPPED
Five moving parts. Only the first two are respectable; the middle two are the ones to be deliberate about; the last one is the moat nobody talks about.
{{ m.n }}
{{ m.name }}
HOW IT WORKS THERE
{{ m.how }}
WHAT IT TELLS US
{{ m.lesson }}
02 · THE NUMBERS I WOULD KEEP IN MY POCKET
Half of these are for the pitch deck. The 31% is the one that matters to us, because it is the crack in their story and the whole reason a keeping app has something honest to say.
{{ s.v }}
{{ s.l }}
Figures as reported in the video note; worth a second source before any of them go in front of an investor.
03 · WHAT TRANSFERS TO CLCTN
Mechanic by mechanic, with a verdict. Steal means do it now; adapt means the shape is right and the incentive is wrong; refuse means it works and we still should not.
MECHANIC
THEIR FORM
OUR FORM
VERDICT
{{ t.k }}
{{ t.them }}
{{ t.us }}
{{ t.v }}
04 · WHERE THE ANALOGY BREAKS
Four differences big enough that copying the Vinted funnel line for line would kill us rather than scale us.
{{ b.k }}
{{ b.t }}
{{ b.b }}
05 · THE ASYMMETRY WE INHERIT
THE 16-POINT GAP
35% feel they own too much. 19% think they buy too much.
Vinted sells a fix for the first number and quietly lives off the second. Emptying a closet makes room, room invites refilling, and the proceeds are already sitting in the app as spendable credit. Their loop is deliberately open at the top.
WHAT A COLLECTOR IS
Someone who has already decided the stock is the point.
A collector does not want the closet emptied; they want it understood. That makes them a terrible Vinted user and an excellent CLCTN user — low churn, high attachment, and a genuine willingness to pay for depth rather than for discharge. Our retention comes from the object staying, not from the next transaction.
Practical consequence for positioning: never claim an ecological benefit we cannot measure inside our own app. Vinted's ecology framing was mostly press-assembled and it is now a liability. If we want a virtue claim, the defensible one is longevity — an object that is catalogued, valued, insured and repaired stays owned. That we can actually instrument: years held per object, repairs logged, objects passed on intact rather than resold blind.
06 · THREE MODELS ON THE TABLE
All three keep cataloguing free and unlimited — that is the Vinted lesson we should not negotiate. They differ in what carries the price.
{{ m.k }}
MY PICK
{{ m.t }}
{{ m.b }}
{{ m.price }}
Risk. {{ m.risk }}
HOW I WOULD SEQUENCE IT
Ship A alone. It needs no counterparty, no logistics, no moderation team, and it prices the thing collectors already ask for out loud. Add B only once a single type has enough depth that a sale inside the app beats a sale on the open marketplace — Vinted almost died precisely by opening six fronts before one of them was solid. Treat C as the long option: every object we catalogue accrues a record with memory, and a marketplace full of amnesiac listings will eventually pay for that. Do not build C on spec.
07 · WHAT I NEED DECIDED
{{ q.k }}
{{ q.t }}
{{ q.b }}