Customer Success

---
name: Customer Success
description: Use when building customer support infrastructure: ticket triage automation, knowledge base systems, workflow automation, or customer health scoring models.
category: Product
version: 1.0.0
tools: []
---

# Customer Success

Build the technical systems that let support teams resolve customer issues efficiently and proactively: ticket triage automation, knowledge base architecture, customer health scoring, and workflow systems that route issues to the right team with the right context. Every support interaction is a signal about the product — the best systems reduce ticket volume by feeding insights back into the product, not just resolving tickets faster.

## Process

1. Ticket intake and classification: accept requests from multiple channels (email, chat, in-app, API), extract structured metadata (account, product area, severity), and apply ML-based classification to assign category, priority, and initial routing.
2. Automate triage: high-severity issues escalate immediately with pager alerts, known issues auto-link to existing incident tickets, password resets and account questions trigger self-service flows, remaining tickets route by product area.
3. Knowledge base architecture: content organized by product area and role, full-text search with relevance ranking, article versioning tied to releases, automated suggestions when tickets match known topics.
4. Customer health score: combine product usage signals (login frequency, feature adoption, API call volume), support signals (ticket frequency, severity distribution, resolution satisfaction), and business signals (contract value, renewal proximity, expansion opportunities) into a composite score predicting churn risk.
5. Escalation management with defined SLAs per priority (P1: 15-minute response, 4-hour resolution; P2: 1-hour response, 24-hour resolution), automated reminders as SLAs approach breach, and escalation paths that notify progressively senior responders.
6. Customer context panel aggregating account details, subscription tier, recent usage, open and recent tickets, known issues, and health score with trend — cutting the time agents spend gathering context.
7. Feedback loop pipeline: identify recurring issues from ticket data, group by root cause, quantify support burden (volume, resolution time, customer impact), generate product improvement recommendations prioritized by impact reduction.
8. Self-service resolution: interactive troubleshooting guides walk customers through diagnostic steps, collect information (error messages, environment, reproduction steps), and either resolve the issue or create a pre-populated ticket with diagnostic context.
9. Automate customer communication: proactive notifications for known issues, scheduled maintenance windows, feature releases relevant to usage patterns, renewal reminders with engagement history.
10. Support analytics dashboard tracking ticket volume trends, resolution time distributions, first-contact resolution rate, satisfaction scores per agent and category, knowledge base deflection rate, and self-service completion rate.

## Standards

- Ticket classification must hit at least 85% accuracy on category assignment; misrouted tickets add latency and frustrate customers and agents.
- Knowledge base articles must be reviewed on every release affecting documented features; outdated articles erode trust more than missing ones.
- Health scores must be computed daily with all signals refreshed; stale scores create false confidence in at-risk account identification.
- SLA timers must account for business hours per customer timezone and exclude weekends/holidays from elapsed time.
- All communication must be logged against the customer record, visible to agents regardless of channel.
- Self-service flows must include an escape hatch to human support at every step; trapping customers in loops that can't solve their problem is a retention risk.
- Analytics must segment by customer tier, product area, and channel for targeted rather than aggregate improvements.

## Verification

- Validate classification accuracy against a labeled holdout set of 500 tickets, confirming category, priority, and routing meet thresholds.
- Confirm SLA monitoring correctly calculates elapsed business hours and triggers alerts at the defined threshold per priority level.
- Test knowledge base search with common question phrasings; confirm the top three results include the relevant article.
- Verify health scores rank known at-risk accounts (recently churned or escalated) lower than healthy accounts in backtesting.
- Confirm self-service flows resolve targeted issue categories without human intervention in at least 60% of attempts.
- Validate the feedback loop pipeline identifies top recurring issues by volume and generates actionable recommendations.