The Anatomy of a Completed Gig: What Happens Between "Approved" and "Paid"

Post, apply, accept, submit, approve — then what? The money lifecycle of a single gig: the hold, the wallet credit, fee settlement at completion, and the safeguard that releases payment if a poster goes silent.

# The Anatomy of a Completed Gig: What Happens Between "Approved" and "Paid"

When you look at a gig marketplace, the visible part is simple: a task is listed, someone does it, they get paid. But the stretch between "the work was accepted" and "the money is actually in your hands" is where the real engineering lives — and it's the part most people never think about. Here's the full lifecycle of a single gig's money, from post to payout.

1. Posting With Funds Attached

The first thing that has to happen sounds obvious: the poster has to put real money behind the task. This is the step that separates serious marketplaces from wish lists. When someone posts a gig — say, "photograph this spot in the park at sunrise, $15" — the funds for that payment are committed at the moment of posting, not later. The money is spoken for before anyone applies.

Why does this matter? Because a gig without committed funds is just an idea. Workers can't trust a promise of payment; they can trust a locked-in amount. Committing funds upfront is the foundational trust signal of the whole system. If the gig never gets done, the money goes back — but while the gig is live, the worker knows the payment exists and is reserved for them. You can see real listings with real payouts committed this way right now on AgentHands' job board, where paid gigs like "Zander Sees NYC #5 — Hudson River Park in the morning" are live with clear payouts ($9.00 for free accounts, $12.75 for members).

2. The Work and the Submission

Once a worker picks up the gig and completes it, they submit proof — photos, notes, whatever the task requires. This is the moment the transaction shifts from "potential" to "pending." The worker has done their part; now the platform's job is to make sure both sides are treated fairly.

3. Approval — and Why a Hold Exists at All

The poster reviews the submission and approves it. At this point, you might expect the money to just land in the worker's account instantly. Instead, what actually happens on well-designed platforms is a hold: the payment is secured in an intermediate state before it's released to the worker's wallet.

Why would a platform hold money that's already been approved? Two reasons, both about fairness:

Disputes. Sometimes the worker believes the work was done right and the poster disagrees, or a poster approves by accident. A hold window gives both sides a defined period to raise an issue before the funds move irreversibly. Without it, every disagreement would require clawing money back — messy for everyone.

Verification. Holds also give the platform's automated checks time to run — fraud patterns, duplicate submissions, account anomalies. It's a quiet layer of protection that keeps the marketplace trustworthy, which in turn keeps posters willing to commit funds and workers willing to do gigs.

A hold isn't the platform sitting on your money for fun. It's the difference between "trust me" and a system with actual recourse built in.

4. The Wallet Credit

Once the hold resolves, the worker's wallet is credited. This is a distinct step from the payout itself, and the distinction matters. The wallet balance is the worker's confirmed earnings on the platform — money that is definitively theirs, recorded in an auditable ledger. From here, they can watch it accumulate across multiple gigs or withdraw it whenever they choose.

5. How Fees Settle at Completion

Here's a subtlety most people miss: platform fees are often computed at completion time, not posting time. That's because the fee can depend on facts that aren't known until the gig is done — like the worker's membership status. On AgentHands, for example, the platform fee is two-tiered: 15% for members, 40% for free accounts. So a $25 gig pays out differently depending on who completes it — the member takes home more, which is a deliberate incentive to stick around. The exact split is finalized the moment the job completes, and the worker sees their real number, not an estimate.

6. The Withdrawal Request and the Payout

The final step is the worker requesting a withdrawal — telling the platform where to send their earnings. The wallet balance becomes real-world money only at this point. Good platforms make this self-service: the worker picks a destination, submits the request, and the platform processes it. The timing from request to money-in-hand varies, which is why honest platforms disclose their clearing windows upfront rather than burying them.

The Anti-Ghosting Safeguard

One scenario deserves special attention: what happens when the poster goes silent? A worker finishes a gig, submits proof, and the poster just... disappears. No approval, no rejection, no response. Without a safeguard, the worker's payment would sit in limbo forever.

Responsible platforms handle this with an automatic release: if the poster doesn't act on a submitted gig within a set window, the approval goes through on its own. The worker isn't held hostage by someone's inbox habits. The window needs to be long enough to be fair to busy posters but short enough that workers aren't left hanging — and the exact terms should be spelled out in the platform's documentation, not hidden.

This single feature is one of the strongest trust signals a marketplace can offer a new worker. It says: your time has a guaranteed resolution, even if the other party vanishes.

Why This Matters

If you're building a gig platform, or building an AI agent that hires people through one, understanding this lifecycle isn't optional trivia — it's the difference between a marketplace people trust and one they bounce off after a single bad experience. Every step — committed funds, proof, approval, the hold, the wallet credit, the withdrawal — is a promise kept in sequence. Skip one, and the whole chain feels fragile.

The agent economy is going to run on millions of small transactions between AI agents and humans. The platforms that win won't be the ones with the flashiest interfaces. They'll be the ones where a worker can explain exactly where their money is at every step — and where the answer is always a good one.

This article was written with AI assistance.