Every few years someone builds a swap site. Offer your design skills, someone offers plumbing, the site matches you up and nobody needs money. It is a lovely idea, people sign up in their thousands, and within six months the listings are tumbleweed. The failure is so consistent it has an economics name: the double coincidence of wants.
The coincidence problem
A direct swap needs four things to be true at once: someone near you wants what you do; that same person does something you want; you both want it now; and you both agree the trade is even. Each condition thins the pool. All four together make matches so rare that the site feels broken even when thousands of willing people are on it. They are all there, staring at each other, unable to trade.
Credits break the coincidence in half
An earned credit separates the two sides of the swap. You do a job for whoever needs your skill, whenever they need it, and bank the value. You spend it with whoever has the skill you need, whenever you need it. Four simultaneous conditions become two independent ones, and the matching problem collapses from near-impossible to merely normal.
Uneven value stops mattering too. A whole-day garden clearance against a twenty minute CV polish is a hopeless direct swap, but through the wallet it is just 8 credits earned and 2 spent, with the difference banked for later.
Why you will never buy credits here
The obvious next feature is letting people top up their wallet with money, and it is the feature that would quietly kill the whole thing, twice over. Commercially, it splits the town into people who do the work and people who buy the work, which is just a worse version of hiring. And practically, a purchasable credit starts to look like stored value, which is the regulated territory of payments law, and rightly so. Earned-only credits keep the exchange what it claims to be: a ledger of favours between neighbours, where every credit in circulation is proof that somebody already showed up and did the work.
What time-banks got right, and where we differ
Community time-banks proved decades ago that earned exchange works, and that it works best locally. Their limitation is the strict hour-for-an-hour rule: an hour of expert bookkeeping and an hour of leaflet folding are not the same value, and pretending otherwise drives skilled members away. TradeOff keeps the earned-only principle and the local boundary, but prices jobs by guide ranges the town itself establishes, so value can be honest without being awkward.
That is the model: earned credits, local wallets, guide prices, human disputes. Simple enough to explain at a bus stop, which is roughly the test a town-scale product has to pass.