Case study · Pegg
A pricing engine for secondhand fashion.
Pegg lets people digitise their wardrobe, see what it is worth, and sell to others who love their style. The hard part is the number: pricing one of a kind, used fashion fairly and instantly.
The challenge
Used fashion has no sticker price.
New retail is easy to price: there is an RRP and a discount. Secondhand fashion is the opposite. Every item is one of a kind, condition varies wildly, and brand and demand move the value far more than the original price tag. Ask a seller to name a number and they freeze, or they guess wrong and the item never sells.
Pegg needed pricing to disappear as a problem. A seller should photograph an item, and the right price should already be there, fair enough that buyers trust it and the seller does not feel short changed.
Our approach
A pricing engine, not a guess.
We built a system that takes an item's brand, category and condition, blends it with live market signals and comparable sales, and applies depreciation modelling to land on a fair price in seconds.
What it does
- Reads brand, category and condition for each listing.
- Matches against comparable sales and live market signals.
- Applies depreciation modelling for age and wear.
- Returns a suggested price the seller can accept or adjust.
The output is explainable, so the number never feels arbitrary. Sellers list without second guessing, and buyers trust that the price reflects the real market.
The outcome
Pricing stopped being the hard part.
The engine ships inside Pegg's listing flow. A seller photographs an item and a fair, defensible price is waiting, in seconds, with no spreadsheets and no guesswork. It removes the single biggest point of friction in putting secondhand fashion up for sale.
Photograph an item, and the right number is already there.
Have a similar problem?
If the maths is the moat, we build it.
Pricing, ranking, recommendation, forecasting. If a decision system sits at the heart of your product, let's talk.