E-commerce
July 1, 2026
Support exceptions are sometimes necessary: late returns, goodwill gestures, special refunds, exceptional replacements, or waived fees. But if they are not documented, they quickly become inconsistent.
Support must handle the exception with empathy while avoiding creating a promise that can be reused by all customers.
This guide shows how to document support exceptions without creating a commercial precedent.
Summary
Why document exceptions?
An exception may be fair in a specific context, but unfair if it is repeated without criteria. Agents need to know why it was granted, who validated it, and what limits apply.
Documentation protects consistency and customer relations.
A well-documented exception remains a controlled decision, not a new hidden rule.
Convert over 2,000 customers on average per month with Qstomy.
The world’s 1st Shopify AI dedicated to customer conversion
Empowering 200+ e-commerce merchants
What information should be noted?
Note the client, order, reason, standard rule, exception granted, proof, amount, validation, date, context, limit, communication sent, and precedent risk.
The record must explain the why, not just the action.
How should the exception be phrased to the client?
Explain the exception as a response to a specific context, without promising that it will always be granted. Avoid phrasing that turns the gesture into a general right.
The customer must understand the solution without seeing it as a permanent rule.
How to set boundaries?
Specify whether the exception is one-time, evidence-based, time-limited, approved by a manager, or non-renewable. Agents need to know when to deny a similar request.
The limit is what makes the exception fair.
If the exception is denied, the documentation must also keep the reason for denial. This record prevents another agent from later granting the same request without seeing the initial context.
How to avoid drift?
Regularly analyze exceptions by reason, agent, amount, and product. If an exception occurs frequently, it may signal a poorly adapted rule, a confusing page, or an operational issue.
Support must transform recurring exceptions into clear policy decisions.
Documentation prevents improvised actions.
Support can also define a validity period for certain exceptions. A return authorization, coverage of fees, or a price match can expire if the customer does not act within the agreed timeframe.
An exception must have a beginning, an end, and a record.
Which flow to follow?
The flow must frame the exceptional decision.
Identify the customer, order, request, standard rule, context, proof, and impact.
Check criteria, validation, amount, history, risk, precedent, and alternative.
Decide, explain the exception, its limit, and what does not change in the rule.
Document, apply the gesture, refuse, transfer, update the policy if recurring.
Measure exceptions, amounts, reasons, repetitions, satisfaction, and margin impacts.
Which examples should be used?
“On an exceptional basis, given the delay noted, we are covering these costs.” “This coverage is linked to this case and does not modify the standard policy.”
The response must be clear and measured.
When to transfer?
Transfer is necessary for a high amount, VIP client, dispute, repeated request, legal risk, out-of-policy exception, social media, or significant margin impact.
The bot must transmit the rule, request, proof, history, amount, validation, and risk.
Which KPIs should be monitored?
Track exceptions granted, reasons, amounts, agents, approvals, recurrences, disputes, satisfaction, and margin.
These KPIs show whether exceptions remain under control.
Which mistakes should be avoided?
Avoid untraceable actions, overly general formulations, exceptions that differ from one agent to another, or the lack of validation for sensitive amounts.
An undocumented exception becomes a risk.
How can Qstomy help?
Qstomy can connect the chatbot to support conversations, tickets, consents, exports, records, proofs, escalation requests, exceptions, business rules, products, and validation procedures to answer accurately.
The chatbot helps the customer understand an export, a record, a product escalation, a denied escalation, or an exception without inventing a document, an agreement, a priority, or a precedent that needs to be verified.
Explore AI support, the AI sales agent, or request a demo.
Key takeaways
Takeaway
Support exceptions must clarify the standard rule, context, proof, validation, amount, limit, communication, and risk of precedent.
What the client needs to understand
The client must receive a tailored solution without turning the exception into a general promise.
The right limit for the chatbot
The chatbot can prepare the case, but it must transfer high amounts, VIPs, disputes, out-of-policy situations, and legal risks.
Enzo
July 1, 2026
