Decide whether and how to escalate a support ticket
intermediateClaude SonnetCustomer ServiceSupportescalationsupport-processtriagecustomer-service
Use case
Use this prompt when a support ticket feels like it might need escalation but you're not sure. Consistent escalation decisions reduce both under-escalation (letting serious issues fester) and over-escalation (overwhelming engineering and leadership with routine requests).
The prompt
You are a senior support manager helping a rep decide whether to escalate a support ticket. Analyze the following ticket and provide a clear escalation recommendation. Ticket details: - Customer name:{{customer_name}}- Account tier:{{account_tier}}(e.g., free, professional, enterprise) - Issue description:{{issue_description}}- How long the issue has been unresolved:{{time_unresolved}}- Customer's expressed urgency or frustration level:{{customer_sentiment}}- What the rep has already tried:{{attempted_solutions}}- Any business impact the customer has mentioned:{{business_impact}}- Renewal date or account health notes:{{account_health}}Evaluate this ticket against these escalation criteria and provide a recommendation: ## 1. Escalation Assessment Rate each dimension on a scale (High / Medium / Low): - **Technical complexity:** Is this beyond tier-1 support capability? - **Customer impact:** How significantly is this affecting the customer's work? - **Account risk:** Is there churn risk, renewal risk, or strategic account concern? - **Time sensitivity:** Is there a hard deadline or compounding urgency? - **Attempted resolution:** Has the rep reasonably exhausted tier-1 options? ## 2. Escalation Recommendation - **Recommended action:** (one of: Handle at tier-1, Escalate to engineering, Escalate to senior support, Escalate to customer success / account management, Escalate to leadership) - **Rationale:** 2–3 sentences explaining the recommendation - **Priority level:** P1 (respond in 1 hour), P2 (respond in 4 hours), P3 (respond next business day) ## 3. Internal Escalation Note A brief, structured note the rep can send with the escalation — summarizing the issue, what's been tried, and what's needed. Under 100 words. ## 4. Customer Communication During Escalation A brief message to send the customer while the issue is being escalated — sets expectations without over-promising.
Variables
{{{{customer_name}}}}Replace with your {{customer name}}{{{{account_tier}}}}Replace with your {{account tier}}{{{{issue_description}}}}Replace with your {{issue description}}{{{{time_unresolved}}}}Replace with your {{time unresolved}}{{{{customer_sentiment}}}}Replace with your {{customer sentiment}}{{{{attempted_solutions}}}}Replace with your {{attempted solutions}}{{{{business_impact}}}}Replace with your {{business impact}}{{{{account_health}}}}Replace with your {{account health}}
Example
Input
customer_name: Harlow Retail Group (account: Josephine) account_tier: Enterprise issue_description: Signal anomaly detection has been sending false positive alerts for 3 days — getting alerts every 15 minutes on a metric that's performing normally; the team has started ignoring all alerts time_unresolved: 3 days customer_sentiment: very frustrated; mentioned that the team has disabled alerts entirely because of the false positives, which defeats the purpose of the feature attempted_solutions: adjusted alert sensitivity threshold — didn't fix it; cleared cache — no change; confirmed data source is updating normally business_impact: team has disabled Signal entirely, so they now have no anomaly detection coverage — the key value driver for their Enterprise upgrade account_health: Enterprise customer, renews in 4 months, had expansion conversation 2 weeks ago
Output
## 1. Escalation Assessment - Technical complexity: **High** — false positive alert behavior after config changes suggests a backend logic or data pipeline issue beyond tier-1 scope - Customer impact: **High** — the customer has disabled the core feature they upgraded for - Account risk: **High** — Enterprise account, renewing in 4 months, expansion conversation recently; this is a direct churn signal - Time sensitivity: **High** — every day they have Signal disabled, they're experiencing zero value from the feature and building a negative narrative around it - Attempted resolution: **High** — standard tier-1 remediation steps have been exhausted ## 2. Escalation Recommendation - **Recommended action:** Escalate to engineering (immediate) + loop in customer success / account management - **Rationale:** The technical issue is beyond tier-1 scope and requires engineering investigation of the Signal alert logic for this specific account. The account risk warrants CSM involvement to manage the relationship through the fix — especially given the renewal timeline and recent expansion conversation. Treating this as engineering-only misses the relationship dimension. - **Priority:** P1 — respond within 1 hour ## 3. Internal Escalation Note Escalating Signal false positive issue for Harlow Retail Group (Enterprise, renewing in 4 months). Customer has received false positive alerts every 15 minutes for 3 days on a normally-performing metric. Standard remediation (sensitivity adjustment, cache clear) has not resolved it. Customer has disabled Signal entirely. This is the primary value driver for their Enterprise plan — needs engineering investigation of alert logic for this account. CSM should be looped in immediately. ## 4. Customer Communication Josephine — I'm escalating this to our engineering team right now, marked as urgent. The false positive behavior you're seeing requires investigation at a level our support team can't access, and I want to make sure the right people are looking at it today. I'll have an update for you within 2 hours — either a fix or a clear explanation of what's happening and a timeline. You shouldn't have to wait days for this.
Tips for best results
- 1The account health field is the most commonly omitted but most critical input. A P3 technical issue becomes a P1 relationship issue when the account renews in 6 weeks.
- 2Over-escalation is a real cost — it overwhelms engineering and trains customer-facing teams not to solve problems independently. Only escalate after tier-1 options are genuinely exhausted.
- 3The internal escalation note format should be standardized across your team. Ambiguous escalation notes that don't include 'what's been tried' waste engineering time.
- 4When in doubt about escalation, the safest move is to loop in a senior support rep for a second opinion rather than escalating all the way to engineering.
- 5After the issue is resolved, run a post-mortem on the escalation: should this have been caught by a tier-1 fix? If yes, update the knowledge base to prevent recurrence.
Related prompts
Draft a response to a customer support ticket
beginnerGenerate a polite, helpful, and complete first-response draft to a customer support ticket, tailored to the issue type and customer tone.
Customer Servicesupportticketresponse
De-escalate an angry customer email thread
advancedGenerate a carefully calibrated response to an angry or hostile customer email that acknowledges their frustration, takes accountability, and redirects toward resolution.
Customer Servicede-escalationangry-customerescalation
Write an internal ticket handoff summary
beginnerGenerate a structured internal summary for handing off a support ticket to a different rep, team, or shift, ensuring nothing is lost in transition.
Customer Servicehandoffinternalsupport-ops
Need help implementing this prompt in your workflow?
Book a call