| name | service-recovery-plan |
| description | Designs service recovery protocols for when things go wrong with response scripts, compensation guidelines, and follow-up procedures. |
| allowed-tools | Read Write Glob |
| metadata | {"author":"matthewhitcham","version":"1.0"} |
Service Recovery Plan
When to Use This Skill
Use this skill when you need to:
- Design proactive protocols for recovering from service failures
- Create response scripts and compensation guidelines for different failure types
- Build follow-up procedures that turn service failures into loyalty opportunities
- Standardize how your business handles things when they go wrong
DO NOT use this skill for individual complaint responses (use complaint-resolution), crisis communication, or business continuity planning. This is for designing the overall service recovery system.
Core Principle
THE SERVICE RECOVERY PARADOX IS REAL — CUSTOMERS WHO EXPERIENCE A FAILURE THAT IS WELL-RECOVERED OFTEN BECOME MORE LOYAL THAN CUSTOMERS WHO NEVER HAD A PROBLEM.
Phase 1: Failure Inventory
Identify what can go wrong and how often it does.
Required Inputs
| Input | What to Ask | Default |
|---|
| Business type | "What product or service do you provide?" | No default |
| Common failures | "What are the top 5 things that go wrong for customers?" | No default |
| Current recovery | "How do you handle things when they go wrong today?" | Ad hoc |
| Compensation budget | "What can you afford to offer when you fail? (refunds, credits, free work)" | Flexible |
| Past incidents | "Describe the worst service failure you have experienced." | No default |
Failure Classification
## Service Failure Types
| Failure Type | Severity | Frequency | Impact on Customer | Example |
|-------------|----------|-----------|-------------------|---------|
| Delivery delay | Medium | [H/M/L] | Inconvenience | Deliverable 3 days late |
| Quality issue | High | [H/M/L] | Rework or dissatisfaction | Deliverable does not meet brief |
| Communication gap | Low-Medium | [H/M/L] | Confusion or frustration | No update for 1 week |
| Billing error | Medium | [H/M/L] | Financial impact | Overcharged or double-billed |
| Complete failure | Critical | [Low] | Major business impact | Service not delivered at all |
GATE: Confirm failure inventory before building recovery plans.
Phase 2: Recovery Protocols
Design specific recovery procedures for each failure type.
Recovery Framework: LAER
## LAER Recovery Model
### L — Listen
Let the customer describe their experience fully. Do not interrupt or defend.
"Help me understand exactly what happened."
### A — Acknowledge
Validate their experience and take ownership.
"You are right — we did not deliver what we promised, and I take full responsibility."
### E — Execute
Deliver a specific fix with a timeline.
"Here is exactly what I am going to do and when: [action] by [date]."
### R — Recover Plus
Go beyond fixing the problem. Deliver unexpected value.
"In addition to [the fix], I am also [additional gesture]."
Recovery Protocol by Failure Type
## Recovery Protocols
### Delivery Delay
**Acknowledge within:** 2 hours of delay becoming known
**Response script:** "I want to be upfront — your [deliverable] will be [X days] later than promised. That is on me. Here is the new timeline: [date]. [Compensation if appropriate]."
**Compensation:** 10-15% discount on current project, or priority scheduling for next project
**Follow-up:** Day of delivery + 3 days after
### Quality Issue
**Acknowledge within:** Same day as reported
**Response script:** "Thank you for flagging this. The quality is not where it should be. I am going to [redo/revise] this and have an improved version to you by [date]."
**Compensation:** Free revision + 10% credit on next project
**Follow-up:** After revision delivery + 1 week later
### Communication Gap
**Acknowledge within:** 4 hours
**Response script:** "You should not have had to chase me for an update. I apologize for the silence. Here is the current status: [update]. Going forward, I will [new communication commitment]."
**Compensation:** Usually not needed — improved communication is the recovery
**Follow-up:** Proactive update within 48 hours
### Billing Error
**Acknowledge within:** Same day
**Response script:** "You are right — there was a billing error. I have [corrected it / issued a refund] immediately. You should see [amount] back in your account within [timeframe]."
**Compensation:** Refund + small credit for the inconvenience
**Follow-up:** Confirm refund received within 3 days
### Complete Failure
**Acknowledge within:** Immediately
**Response script:** "I need to be honest with you — [what happened]. There is no excuse. Here is what I am doing: [full remedy]. [Significant compensation]."
**Compensation:** Full refund + redo at no cost OR full refund + additional credit
**Follow-up:** Personal call within 48 hours + 1-week check-in + 30-day check-in
GATE: Present recovery protocols for review.
Phase 3: Compensation Guidelines
Define what to offer and when.
Compensation Matrix
## Compensation Guidelines
| Severity | Our Fault (100%) | Shared Responsibility | Minor Inconvenience |
|----------|-----------------|----------------------|-------------------|
| Low | Apology + expedited fix | Apology + fix | Apology |
| Medium | Fix + 10-20% credit | Fix + 10% credit | Fix + goodwill gesture |
| High | Fix + 25-50% refund | Fix + 15-25% refund | Fix + 10% credit |
| Critical | Full refund + redo free | Full refund | Partial refund + fix |
Compensation Rules
- Always fix the issue FIRST, then discuss compensation
- Offer compensation proactively — do not wait for the customer to demand it
- Never offer more than the total project/subscription value
- Monetary compensation for monetary impact; time compensation for time wasted
- Document all compensation for financial tracking
Phase 4: Prevention and Measurement
Use recovery data to prevent future failures.
Failure Prevention
After each recovered incident, document:
## Post-Incident Review
**Failure:** [What happened]
**Root cause:** [Why it happened]
**Recovery executed:** [What was done]
**Customer outcome:** [Retained / Lost / Upgraded loyalty]
**Preventive action:** [What changes to prevent recurrence]
**Owner:** [Who implements the change]
**Deadline:** [When]
Recovery Metrics
## Service Recovery Dashboard
| Metric | Target | Current |
|--------|--------|---------|
| Mean time to acknowledge | Under [X hours] | — |
| Mean time to resolve | Under [X days] | — |
| Recovery satisfaction rate | 80%+ | — |
| Customer retention after failure | 85%+ | — |
| Repeat failure rate (same type) | Decreasing | — |
Quarterly Recovery Review
- How many service failures occurred?
- What types are most common?
- What is the retention rate after recovery?
- Are preventive actions reducing failure rates?
- Is the compensation budget on track?
Anti-Patterns
- Denying the failure — customers know when something went wrong. Denial insults their intelligence and kills trust.
- Over-compensating — giving away the farm for a minor delay trains customers to expect excessive compensation.
- Under-compensating — a 5% discount for a catastrophic failure is insulting. Match the recovery to the impact.
- No follow-up — fixing the problem without following up leaves the relationship in limbo. Close the loop.
- Same failure repeating — if the same failure happens 3+ times, the recovery plan is not the problem — the process is.
Recovery
- User has never built a recovery plan: Start with the single most common failure type. One protocol is better than none.
- Customer refuses the recovery offer: Ask what would make it right for them. Sometimes it is not about compensation — it is about being heard.
- User cannot afford monetary compensation: Offer non-monetary recovery: priority service, personal attention, free consultation time, early access to new features.
- Recovery failed and customer is leaving: Conduct a graceful exit. Ask for feedback, apologize sincerely, and leave the door open. Some customers return later.
- Team does not follow recovery protocols: Role-play scenarios during team meetings. Make the protocols accessible (not buried in a document no one reads).
