How much can a bonder earn on Pieces?

The kickback on a funded Piece is the kickback percentage times the raise, split pro-rata across the bond pool. The pool is capped at that same percentage of the funding goal, so what a stake earns depends on how full the pool ends up: the fewer bonders sharing the kickback, the larger each share. On a $500 Piece with a 20% kickback, a $50 stake in a full $100 pool settles at $100, and the same stake alone in the pool settles at $150.

Is there a limit to how much I can bond?

Yes, and it is not arbitrary. The bond target is the kickback percentage times the funding goal, so a $500 goal with a 20% kickback accepts exactly $100 of bonds. Once the pool hits that cap, further bonds are rejected. Bonding also closes permanently the moment a Piece funds.

What happens to bonders if a campaign almost funds?

They lose everything. A Piece that reaches 92% of its goal at the deadline has failed: it stays sealed, the entire bond pool is forfeited, and backers are refunded in full plus that pool split pro-rata. There is no partial credit for a near miss, because a bond that paid back part of itself would not be worth anything as a signal.

Is there any case where a failed Piece returns the bond?

One. If a Piece ends with no backing at all, there is nobody to pay a refund bonus to, so bonders get their full stakes back. Any backing at all, even a single dollar, means the whole bond pool goes to the backers.

Four Stakes, Four Endings: A Bonder's Field Guide

A Piece caps its bond pool at the kickback it pays out and splits that kickback pro-rata across whoever staked. Four scenarios with real numbers, including the one ending where bonders get their stakes back.

We opened this series with why bonding exists at all. Now the part that decides what you actually walk away with, because the same stake on the same Piece has four very different endings.

The Piece

Theo is a photographer. He locks a 40-image photo essay on Pieces with two numbers that matter:

Funding goal: $500
Kickback: 20%

Those two numbers set a third one automatically. The bond target is the kickback percentage times the goal, so this Piece accepts exactly $100 of bonds and not a dollar more.

That is the fact worth internalizing before any of the scenarios, because it is what makes the math clean. If the Piece funds, the kickback paid out is 20% of $500, which is $100: precisely the size of the full pool. The pool is capped at exactly what it stands to earn.

Ending one: the pool is full

You stake $50. One other person stakes $50, and the pool is full at $100. Backers see a fully bonded Piece, the goal is hit, and everything settles.

You hold half the pool, so you earn half the kickback: $50. You put in $50 and take out $100.

Your $50 came back as $100: the stake, plus a kickback share the same size. That is not a lucky configuration. The cap and the kickback on this Piece are both $100, so the full pool split a kickback equal to itself. Theo, meanwhile, takes the $500 raise minus the $100 kickback minus the 3% platform fee on his share, which is $388, and the essay is public for everyone.

Ending two: the pool is underfilled

Same Piece, but this time nobody joins you. You stake $50, nobody else bonds, and the Piece funds with the pool sitting at half its cap.

The kickback is still $100, because it is a slice of the raise and not of the pool. You are the only one holding a claim on it, so all $100 is yours. You staked $50 and take out $150.

The pool can never exceed the kickback it pays out, so the fewer stakes sharing that kickback, the larger every share.

Before that reads like an exploit, look at why the pool was empty. Nobody else would take this bet. You were not paid extra for being clever about timing; you were paid extra for carrying risk alone that a crowd declined to share. And you cannot lie in wait for it, because of the rule in the next scenario.

Ending three: you cannot get in

You find a Piece from a creator you love, and you cannot stake a dollar. Two doors do this.

The first is the cap. The pool is already at its target, so the contract rejects your money. A full pool is not a snub, it is information: people with money at risk stood behind this until it stopped accepting them.

The second is more interesting. Bonding closes the instant a Piece funds. Once the goal is hit, the outcome is settled and the bond window is gone forever. So the strategy of waiting until a Piece obviously will succeed and then bonding it cheaply does not exist. The moment a bet becomes safe, it is no longer available. Every bond ever placed on Pieces was placed while the ending was genuinely unknown.

If you are locked out, do the thing the pool is counting on. Back it. Backing is what actually pushes a Piece to its goal, and it is the one position that comes out ahead either way.

Ending four: the near miss

Last one. Your $50 is in a full $100 pool, the campaign runs hot, and at the deadline the backing sits at $460 of the $500 goal.

Ninety-two percent is a failure. The essay stays sealed forever, the entire $100 pool is forfeited, and the backers are refunded in full plus that pool spread across their $460, about 21 cents of bonus for every dollar they backed. They had a good day. You lost all $50, at 92%, and it would have looked the same at 5%.

That should sting, and it is the most important paragraph here. There is no partial credit because partial credit would rot the signal. A bond that returns most of itself on a near miss is barely a bet, and a pool full of hedged half-bets tells backers nothing. Your stake is meaningful to everyone watching precisely because the mechanism will actually take it.

There is exactly one exception, and it is a strange one. If a Piece ends with zero backing, nobody is owed a refund bonus, so the bonds go back to the bonders. Total indifference returns your stake; a campaign that tried and fell short does not.

What the four endings add up to

The kickback is fixed by the Piece; the pool decides your share of it. Get in alone and the share is larger, because nobody else would take the bet. Arrive late and you may find the door shut, by the cap or by the funding itself. Miss by 8% and you are out everything, which is the price of the signal being worth anything at all.

None of this rewards clever timing, and that is deliberate. It rewards being right about whether a creator’s audience will show up. The rules are in what a bonder is, the theory behind them in the dominant assurance contract, or go find a Piece whose audience you already understand.