E-commerce
June 28, 2026
Every support decision creates a precedent: accepting a late return, offering a gesture, refusing a warranty, reimbursing fees, or extending a promotion. If these decisions are not documented, the team loses consistency.
The chatbot can help collect the context and summarize the decision, but sensitive exceptions must still be validated. Good documentation protects the customer, the agent, and the margin.
This guide shows how to document support decisions to avoid costly exceptions and contradictory answers.
Summary
Why document decisions?
A customer may return several days later and speak to another agent. Without a clear record, they must repeat their story, and the new response may contradict the first.
Documenting the decision makes it possible to understand what was requested, verified, granted, or denied, and why.
An undocumented support decision quickly becomes an exception that is impossible to explain.
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
Which decisions should be followed?
It is necessary to document refunds, credit notes, discounts, commercial gestures, refusals, guarantees, late returns, delivery fees, replacements, carrier disputes, and B2B exceptions.
Simple FAQ responses do not require the same level of detail as financial or relational decisions.
What elements should be noted?
The note must contain the issue, the evidence, the consulted rule, the decision, the amount, the approver if necessary, the message sent to the customer, and the next action.
It must be short but sufficient for another agent to take over without re-doing the entire investigation.
How to avoid hidden exceptions?
An exception must be named as such. If it is linked to a promise, a strategic client, a brand error, or a very specific situation, this must appear clearly.
Otherwise, it can be wrongly repeated for cases that are not comparable.
How can the decision be made understandable to the client?
The client does not need to see all the internal logic, but they must understand the general rule and the reason for the decision. A refusal without explanation increases follow-ups.
The bot can help formulate a clear, complete, and respectful response.
This clarity is also useful internally: it prevents an agent from granting a new exception because they do not understand the previous decision.
Sensitive decisions must remain easy to find, especially when they involve an amount, a guarantee, or a promise made to the client.
Which flow to follow?
The flow must keep an actionable record.
Identify the request, order, customer, evidence, applicable rule, and impact.
Distinguish between standard decisions, exceptions, goodwill gestures, refusals, manager validations, or escalations.
Document the decision, amount, reason, source, approver, and message sent.
Update the ticket so the next agent understands what to do next.
Analyze recurring exceptions to correct policies, products, or journeys.
Which messages should be used?
For internal tracking: "Decision: credit note of 10 euros granted after delay confirmed, according to delay gesture rule."
For customer: "In this case, we can offer a credit note because the delay is confirmed by the tracking."
For refusal: "The request cannot be accepted because the time limit provided by the policy has been exceeded."
When to transfer?
Escalation is necessary if the decision exceeds the authorized threshold, if a customer disputes, if an exception is requested, if a significant amount is at stake, or if the rule is ambiguous.
The bot must transmit the request, evidence, rule, proposal, amount, history, and risk.
Which KPIs should be monitored?
Track exceptions, gestures, contested refusals, manager validations, amounts granted, contradictory decisions, and reopenings linked to an insufficient score.
These indicators show whether the team is making consistent decisions.
Which mistakes should be avoided?
Avoid leaving a decision without a reason, hiding an exception, documenting too late, or noting unnecessary personal information.
Documentation should support the continuity of care, not become a confusing archive.
How can Qstomy help?
Qstomy can connect the chatbot to business rules, orders, packaging, support decisions, payments, downloadable products, and escalation procedures to answer clearly, then transfer sensitive cases with an actionable summary.
The chatbot helps the customer move forward without inventing a discount, a package privacy guarantee, a support exception, a double charge refund, or download access that has yet to be confirmed by a reliable source.
Explore AI support, the AI sales agent or request a demo.
Key takeaways
Key Takeaways
Support decisions must document the rule, proof, decision, amount, approver, customer message, and next action.
What the customer must understand
The customer must receive a consistent response, even if they speak to a different agent.
The right limit for the chatbot
The chatbot can summarize decisions, but it must hand over exceptions, high amounts, ambiguous rules, and disputes.
Enzo
June 28, 2026
