Purpose
Get serious or stuck customer issues to the right person quickly, with enough information that nobody has to ask the customer to repeat themselves, and keep the customer informed until it is fixed. This procedure sets when to escalate, to whom, and what happens after.
Scope
Covers tickets raised by customers through [email, chat, phone or portal] that front-line support cannot resolve. Does not cover security incidents or data breaches, which follow the incident response SOP, or billing disputes, which go to finance.
Roles and responsibilities
| Role | Responsibility |
|---|---|
| Support agent (tier 1) | Triages, sets severity, resolves or escalates, keeps the customer updated |
| Specialist (tier 2) | Investigates escalated issues and resolves or escalates further |
| Support lead | Owns this SOP, handles complaints and customer escalations, reviews trends |
| [Engineering or product on-call] | Fixes product defects and outages |
| [Account manager] | Informed on any escalation for their accounts |
Before you start
- Access to [helpdesk], [status page] and [chat tool] escalation channel
- The severity table below
- The on-call rota at [link]
- Customer’s contract or plan level, if response times vary by plan
Severity levels
| Severity | Meaning | First response | Update customer every |
|---|---|---|---|
| 1, critical | Service down or unusable for many customers, or data at risk | [30 minutes] | [1 hour] |
| 2, high | Key feature broken for a customer, no workaround | [2 business hours] | [4 business hours] |
| 3, normal | Problem with a workaround, or a single user affected | [1 business day] | [2 business days] |
| 4, low | Question or minor issue | [2 business days] | On resolution |
Set these times to match what you have promised customers in contracts or on your website.
Procedure
- Triage the ticket. Support agent reads the ticket, sets severity using the table, and checks for a known issue or existing ticket from the same customer.
- Try to resolve. Support agent follows the relevant help articles and SOPs. Escalate when any of these is true:
- The fix needs access or knowledge tier 1 does not have.
- The issue is severity 1, or several customers report the same problem.
- The first response or resolution target will be missed.
- The customer asks for a manager, or mentions cancelling or legal action.
- Write the handoff note. Support agent adds an internal note before escalating:
- What the customer reported, in their words
- Steps to reproduce, and what you already tried
- Account, plan, environment and screenshots or error messages
- Severity and why
- What you have told the customer
- Escalate. Support agent assigns the ticket to the tier 2 queue. For severity 1, also post in [escalation channel] and page [on-call] using [paging tool]. Complaints and manager requests go to the support lead.
- Tell the customer. Support agent tells the customer it has been escalated, who is handling it, and when they will hear next. Do not promise a fix time you do not know.
- Acknowledge the escalation. Specialist accepts the ticket within the first response time for its severity and becomes the owner.
- Investigate and resolve. Specialist works the issue, escalating to [engineering] with a linked bug report if it is a product defect.
- For severity 1, the support lead posts on [status page] and updates it at the set interval until resolved.
- Update the customer on schedule. Ticket owner updates the customer at the interval in the table, even if the update is “still working on it”.
- Confirm the fix. Ticket owner explains what was done and asks the customer to confirm it works. Close the ticket after confirmation or after [3 business days] without a reply.
- Record the cause. Ticket owner tags the ticket with the cause and one prevention action: a fix, a help article or an SOP change.
Checklist
- Severity set using the table
- Handoff note complete before escalating
- Customer told it was escalated and when they will hear next
- Escalation acknowledged within the response time
- Updates sent on schedule
- Fix confirmed with the customer
- Cause and prevention action recorded
Exceptions and escalation
- Severity 1 outside business hours: Page [on-call] directly. The support lead is informed but does not need to approve.
- No response from tier 2 within the response time: Escalate to the support lead.
- Possible security or data issue: Stop, do not share details with the customer yet, and follow the incident response SOP.
- Customer threatens to cancel or take legal action: Support lead and [account manager] take over the conversation.
Records to keep
- The full ticket history, handoff note and customer updates in [helpdesk]
- Linked bug reports in [issue tracker]
- Status page history for severity 1 incidents
- Monthly escalation review notes in [shared drive]
Review
The support lead reviews escalations monthly for patterns and reviews this SOP every 6 months, after every severity 1 incident, and whenever response time commitments change.