When customers cannot complete purchases or orders are not reaching fulfillment, start with the specific failure and its impact. A lower conversion rate may be a useful signal, but it does not identify the cause. Separate a reporting problem from a failure that customers are experiencing.
A working incident plan names who will investigate, who can make changes, how customers will receive updates, and what evidence is required before calling the issue resolved. A small store can assign several of these responsibilities to the same person.
Confirm and describe the problem
Record the time, affected market, device, product, and step in the customer journey. Save the error message and relevant order references. Avoid collecting payment details or unnecessary personal information in a shared incident document.
Try to reproduce the failure through the appropriate test process. Compare what the customer sees with the admin, payment provider, fulfillment system, and reporting tools. A successful payment followed by a missing confirmation email is different from a declined payment, even if both customers describe checkout as broken.
Check Shopify's status page when investigating a Shopify issue. A reported platform incident can inform the investigation, but an operational status does not establish that your own store is working. Shopify notes that issues affecting a small proportion of stores may not appear there.
Limit harm while preserving evidence
Identify whether customers can place orders that you cannot fulfill, whether an incorrect price is visible, or whether an integration is creating duplicates. Use the controls appropriate to the actual failure, and keep a record of the changes you make.
Avoid changing several unrelated systems at once. Record a timestamp, reason, and expected result for each intervention so the investigation can distinguish the cause from a later side effect.
For a suspected account compromise, use the provider's recovery process and involve the person responsible for security. For a delivery or supplier incident, focus on affected products and orders. These require different responses from an analytics outage.
Give customers a useful update
Explain the confirmed impact in plain language, what the customer can do now, and when you will provide another update. Do not promise a resolution time that the team has not established.
For delayed orders, provide the relevant order-specific information through a suitable private channel. A public notice should not expose customer names, addresses, or order histories. Keep support wording consistent so customers do not receive conflicting instructions.
Update the message when the impact changes. A short, accurate update is more useful than repeatedly saying that everything will be fixed soon.
Verify recovery across the actual journey
Use a checklist based on the failure. For checkout, include payment, order creation, stock changes, shipping rules, confirmation messages, and the handoff to fulfillment. Shopify documents test-order methods and their limits. Follow the appropriate method and do not leave a live payment provider in test mode; customers cannot make live payments through a provider left in test mode.
Check the affected market and device, not only the administrator's usual environment. Confirm that duplicate or failed records have a reconciliation plan. A page loading again does not prove every affected order is accounted for.
For restored data, verify that the backup contains what you need and that you can use it. The NCSC's current backup guidance emphasizes knowing how to restore and checking the contents of a backup.
Turn the incident into a maintained procedure
Write a short review with the timeline, confirmed cause, impact, recovery evidence, and outstanding actions. Keep uncertain explanations separate from verified findings. Assign an owner and review date for each corrective action.
Update the runbook and contact list while the details are still available. Test recovery procedures at a frequency suited to how often your store changes and how much data it can afford to lose. The objective is a repeatable response that protects customers and establishes recovery with evidence.





