#1424 2016 · Too Good To Go · Food surplus marketplace
Restaurants sell leftovers at a third of value — as long as no one knows what's inside
the problem
Day-end surplus can't be marked down without cannibalizing tomorrow's full-price sales
background
Bakeries, cafes and buffets end each day with unsold food that costs the same whether sold or binned. Deep markdowns fail for a known reason: the same customers who would pay full price tomorrow buy the discount tonight, so clearance competes with the menu, and most surplus ends up in the bin.
Too Good To Go — built in Copenhagen by late 2015, saving its first meals in March 2016 — turned the markdown into a sealed mystery: surprise bags filled with whatever the store has left, sold at roughly a third of the food's value and collected at closing time. Businesses pay a small annual fee plus a per-bag fee — about 39 euros a year plus per-meal fees in Europe, $1.79 a bag and $89 a year in the US.
what everyone would do
Slash prices 70 percent at closing time — and watch regulars stop buying at 5pm to buy at 8pm; the discount you needed to clear stock starts setting your reference price.
what they saw
The markdown failed not because it was too small but because it was too visible. Make the discount a mystery and the full-price buyer cannot wait for what they cannot choose — clearance stops competing with the menu.
the move
The surprise is the mechanism. Because the customer cannot choose the contents, the bag cannot substitute for a planned full-price purchase — it sells the store's marginal leftovers to flexibility-tolerant customers without teaching anyone to wait for the discount. Bags typically cost $3.99-9.99 in the US; pickup happens at closing with staff already present; operations research has since formalized the pattern as 'surprise clearance', finding it can raise store profits while cutting waste.
why it works
Cannibalization requires substitutability: a half-price croissant is the same croissant you would sell tomorrow, but a sealed bag of whatever is left is not a substitute for anything specific, so full-price demand is protected. Customers self-select into those flexible about contents at a third of the price; closing-time pickup uses staff who are present anyway; scarcity and a countdown make clearance a small treasure hunt that brings users back daily. The platform's per-bag fee aligns it with volume rather than listing counts.
the payoff
37,500 stores by 2021 and ~100M users across 17-19 countries later; 141M bags sold by June 2022 (company figures); revenue ~$162M in 2023
where it breaks
The mystery hides quality: one stale or skimpy bag burns a customer who blames the store, and dietary needs only partly filter through food-type tags. It works only for genuinely perishable surplus — masking a discount on durable goods just creates a discount channel. Sparse store density makes pickup windows wasted trips, and by the company's own figures the operator has yet to turn a full-year profit on its per-bag fees.
what came after
Surprise bags spread across surplus-food apps and in-store programs; supermarkets and hotel chains adopted end-of-day mystery boxes; the peer-reviewed operations literature now treats surprise clearance as a named pricing strategy.
references
- [1]Too Good To Go (Tools of Change case study)Tools of Change, 2022toolsofchange.com
- [2]How Too Good To Go helps people find leftover food at huge discountsCNBC, 2024cnbc.com
Widely retold, only partly documented. Filed as hearsay.