How refund decisions are actually made
Most refund advice tells you what to do. This one flips the desk around: what does the person on the other side actually weigh when they approve or refuse you? Understand how the decision is made and your claim almost writes itself — because you start handing the agent exactly what they need to say yes.
The agent is looking for a reason to approve — that they can defend
A support agent is not your enemy, but they are accountable. Approve too freely and they answer for it; refuse a genuine case and they generate an escalation. What they want is a claim they can approve and justify in one line: clear fault, in the record, reasonable ask. Everything that helps you is really about making their approval easy to defend.
What they can actually verify
An agent can see things you cannot: the delivery timeline and GPS track, the "delivered" timestamp and proof photo, the itemised bill, your account history, and sometimes the rider's notes. They cannot see inside your kitchen. So claims they can check against their own data — "marked delivered at 7:52 but the map pin is two streets away" — are far stronger than claims that rest only on your say-so. Point at what is in their system, and you are arguing on their turf.
The three things that move a decision
- Specificity — a named item and a named fault is a data point; "my order was bad" is a mood. Agents action data points.
- Timing — a complaint filed inside the window, before you marked the order received, is one they can still verify. A day later, they often cannot.
- Proportion — asking for exactly your loss reads as honest; asking for the whole tab over one side flags the account for review.
What quietly counts against you
Decisions are not made in a vacuum. A history of frequent, large, or vague claims makes the next one harder — not because any single case is weak, but because the pattern invites caution. This is the real reason the discipline in our other guides matters: honest, proportionate claims keep your account in good standing, which makes every future claim easier to win.
Why this connects to escalation
If a first agent still refuses a clean case, it is often a script or a cautious call, not a final answer — which is why our escalation ladder works: a supervisor re-reads the same strong facts with more authority to say yes.
Practise thinking like the desk
The Wrong Order is, underneath, a simulation of exactly this decision: a fictional agent and an independent board weigh your case on specificity, timing, and evidence, and reward the claim that is easy to justify. Playing it is a fast way to internalise what the real desk is looking for.