Play It Forward: designing a growth loop out of generosity

TL;DR

Play It Forward is Mini Marketa's donation program: families hand over outgrown toys, receive store vouchers, and, crucially, host donation parties that bring their friends into the loop too. It solves the marketplace's existential problem (supply) while acquiring customers at the same time. The interesting product work sits in the incentive structure, the trust design, and the operational machinery that keep a generosity loop running week after week.

The problem: marketplaces die on the empty side

A curated secondhand marketplace has a chicken-and-egg problem with a twist: supply is the hard side, and it arrives in physical bags that someone has to collect, sort, and photograph. Buying stock outright kills margins. Waiting for organic donations doesn't scale. I needed a system where supply wanted to come to me, repeatedly, and brought demand along with it.

The loop

1. A family signs up to donate a bag of outgrown toys (solo, or as a party host who gathers bags from friends).
2. I collect on a planned pickup run.
3. The donor gets a store voucher: $25 for hosts, $10 for participating friends.
4. Vouchers convert donors into first-time customers; their outgrown toys become stock; their friends become new contacts who've now had a warm first touch with the brand.
5. Purchases fund the program, and toys that can't be sold flow to a charity partner, with a live donation counter on the storefront making the loop visible.

Each cycle grows all three sides: inventory, customers, and audience.

Incentive design: the details carry the economics

Host/friend asymmetry. Early versions rewarded everyone equally. The redesign pays the host meaningfully more ($25 vs $10), because hosts do the organising work and each party multiplies the bags collected per pickup trip. My scarcest resource is collection time, so the incentive targets the behaviour that leverages it.

Store credit, not cash. Vouchers cost me margin rather than revenue, and they guarantee a first purchase. A donor's first order is the moment the loop closes.

Bounded generosity. Voucher values, what qualifies as a bag, and repeat-donation rules are all explicit. Generosity programs without edges get quietly exploited, and resentment kills community programs faster than fraud does.

The operational machinery

A community program lives or dies on feeling personal while running on rails:

• Signup processing: every signup email is parsed, tracked in a live sheet, answered with a warm confirmation, and labelled, semi-automated so nothing falls through.
• Pickup run sheets: collections are batched into planned routes, and a script builds the day's run sheet from the tracker.
• Locked message copy: confirmations and voucher emails use tested, consistent copy, so the tone stays right even when the sending is automated.
• Partner channels: the same loop extends to institutions, like a childcare-centre collaboration that turns a toy drive into a pickup point and reaches a whole community of parents in one relationship.

Results and lessons

The program is now the marketplace's primary supply channel and its cheapest customer acquisition.

Design the incentive around your scarcest resource. Mine was collection time, so the money points at hosts who aggregate bags. Most incentive schemes fail by rewarding volume of participants instead of the leverage behaviour.

A growth loop needs product, process, and promise. The voucher is the product, the run sheets are the process, and showing donors where their toys end up is the promise. Drop any one and the loop stalls.

Community programs are trust products. Every automated email had to sound like me on my best day, because for most donors it is the relationship.