Coffee customer service - delivery spill complaint checklist - four questions to size the loss, three compensation tiers, and a same-cup-type recurrence review
You are the delivery-complaint supervisor of a coffee shop.
Turn this spill-complaint checklist into a printable
station version: keep the three-part structure, leave the
tier amounts and authority as [ ] for me to set, and lock
everything else.
[Step 1: Four sizing questions (ask all four at intake;
missing one means no tier assignment)]
1. Which channel: platform order (use the platform's
damage channel) / self-delivered or runner order
(the store handles it)
2. Which drink: name + cup type + hot or iced (decides
which cup type the review line files under)
3. Photos: spill severity is judged from photos; where
the customer's description and the photo disagree,
judge by the lighter one
4. Drinkable or not: log the customer's own judgment as
stated; never ask them to "pour it out and measure
what's left" - that request only upgrades a small
spill into a bad review
[Step 2: Three-tier compensation (tier by photo, not by
inquiry into cause)]
- Tier 1, rim seepage: drink essentially intact, less
than a mouthful out - remedy: apology + [compensation
form, e.g. coupon]
- Tier 2, clear spill: visible spillage, volume
obviously short - remedy: [partial refund / remake,
one, customer's choice]
- Tier 3, undrinkable: cup dumped or contaminated -
remedy: full refund or rush remake, customer's choice
Authority note: any tier-three payout beyond the agent's
authority goes straight to the manager - no haggling with
the customer; platform orders go through the platform's
damage channel first, with the store's rubric covering
only the gap the platform doesn't
[Step 3: Filing and the review line (mandatory per
complaint; "handled" is not "finished")]
- Ledger entry: date / channel / cup type / tier /
remedy / delivery mode (platform courier / runner /
own delivery)
- Review trigger: the same cup type plus the same spill
pattern at least [3] times in one week escalates to
the manager, with three fixed checks - lid-to-rim
fit, foam headroom below the rim, sealing and carrier
used per standard
- The review conclusion goes back into the ledger; if
the same cup type triggers again the following week,
note "checked, recurring" and hand the manager the
swap-cup-or-swap-lid decision
Handling discipline:
- No cause-language toward the customer like "the
courier hit a bump" - root-causing is an in-store
activity; toward the customer there is only resolution
- Never offer compensation in exchange for deleting a
review; public review replies follow a separate
rubric, not this checklist
- No photo means tier 1 automatically, noted "no photo"
- do not argue tiers with the customer
Formatting rules:
- Three sections, one per step; items within a step
ordered by execution
- All compensation-related [ ] collected into a "store
compensation rubric" table at the page foot so one
edit applies everywhere
- No "raise service awareness" summary section
Why it is written this way
The biggest enemy of tiered compensation is "the standard varies by agent": the same one-third-spilled latte gets half refunded by the morning agent and fully comped by the evening one, customers compare notes, and the shop is now "sizing people up" - the most common path from complaint to bad review. This rubric anchors tiers on the photo, the one objective artifact both sides accept, and rules that description-photo conflicts resolve to the lighter reading: consistency is worth more than any single payout's margin.
The review line is the watershed between this checklist and "pay and forget" handling. Coffee spills have strong physical causes - a cup type paired with the wrong lid, foam textured too full, a sealer at the wrong temperature - and any of these makes the same drink spill week after week while the shop pays for each occurrence without fixing anything. The "same cup type 3+ per week" trigger, rather than "when the agent feels it's a lot," removes the dependence on personal alertness; the three fixed checks keep the escalation itself procedural.
"No courier-bump talk" exists because cause-language reads as blame-shifting to the customer and is often simply wrong - charging a lid mismatch to the courier pays money and fixes no cup. Root-causing stays inside the ledger (delivery mode is a field precisely so spill clustering by delivery type can be seen later); the customer-facing output is resolution only. Banning "pour it out and measure" is a pure field lesson: asking the customer to perform a measurement upgrades a tier-1 spill into a tier-2 review.
Variables
| Variable | Notes |
|---|---|
| Compensation rubric table | The concrete remedy and amount per tier, centralized for one-point updates |
| Review trigger | Spills of the same cup type per week before escalation; set by order volume |
| Agent authority | Which tier-three payouts the agent may decide alone; beyond goes to the manager |