Real Examples

Real SOP Examples

These examples show the 7-section framework in action. Each follows Level 2+ detail. Copy and customize for your organization.

Sales

Example 1: Process Inbound Sales Lead

A per-event SOP triggered by every new lead form submission. Demonstrates BANT scoring, conditional routing, and multi-path steps.

# Process Inbound Sales Lead

## 1. Metadata
- **Owner:** Sales Representative
- **Frequency:** Per-event
- **Trigger:** New lead form submission, email inquiry, or referral notification
- **Estimated Time:** 15 minutes
- **Automation Level:** Level 2

## 2. Purpose
**Why this exists:** Route and qualify inbound leads within 2 hours of receipt to maximize conversion rates.

**What happens if not done:** Leads go cold. Response time is the #1 predictor of conversion - leads contacted within 5 minutes are 21x more likely to convert.

**Success criteria:** Lead is scored, categorized, assigned to correct pipeline stage, and first contact is made within 2 hours.

## 3. Prerequisites
**Tools needed:**
- [ ] CRM (HubSpot / Salesforce) - [CRM URL]
- [ ] Email client - [Email URL]
- [ ] Lead scoring rubric - [Internal doc URL]

**Permissions required:**
- [ ] CRM write access (Contacts, Deals)
- [ ] Email send permissions

**Information needed:**
- [ ] Lead form submission data
- [ ] Current pipeline capacity

## 4. Steps

### Step 1: Check lead source
**Action:** Open CRM notifications or email inbox. Identify the lead source (website form, email inquiry, referral, event).
**Expected result:** You can see the lead's name, email, company, and source.
**If something goes wrong:** If form data is incomplete, check the website form submission logs.

### Step 2: Review lead data
**Action:** Open the lead's form submission. Note: company name, role/title, company size, stated need, how they found us.
**Expected result:** Clear picture of who this person is and what they need.

### Step 3: Score the lead (BANT)
**Action:** Apply BANT framework:
- Budget: Can they afford the solution? (Check company size, funding stage)
- Authority: Are they a decision maker? (Check title)
- Need: Do they have a problem we solve? (Check stated need)
- Timeline: Are they buying soon? (Check urgency signals)

Score each 0-25. Total score out of 100.

**Decision point:**
- IF score >= 80 → Hot lead. Go to Step 4a.
- IF score 50-79 → Warm lead. Go to Step 4b.
- IF score < 50 → Cold lead. Go to Step 4c.

### Step 4a: Hot lead handling
**Action:** Create CRM contact → Set pipeline stage to "Qualified" → Draft personalized email referencing their specific need → Schedule follow-up for next business day → Notify sales manager via Slack.
**Expected result:** Contact created, email sent within 30 minutes, follow-up scheduled.

### Step 4b: Warm lead handling
**Action:** Create CRM contact → Set pipeline stage to "Nurture" → Add to automated nurture email sequence → Schedule manual check-in for 1 week.
**Expected result:** Contact in nurture sequence, check-in scheduled.

### Step 4c: Cold lead handling
**Action:** Create CRM contact → Set pipeline stage to "Newsletter" → Add to monthly newsletter list → No immediate follow-up.
**Expected result:** Contact created, added to newsletter.

### Step 5: Log activity
**Action:** In CRM, log: lead source, BANT score, pipeline stage assigned, first action taken, notes on conversation.
**Expected result:** Complete activity log visible to entire sales team.

## 5. Decision Trees

### Duplicate lead check
IF lead email already exists in CRM
  THEN check last activity date
  IF last activity > 90 days → treat as new lead, update existing record
  IF last activity < 90 days → notify original owner, do not re-assign

### Competitor inquiry
IF lead is from a known competitor (check against competitor list)
  THEN do NOT add to pipeline
  Log as "Competitor Inquiry" in CRM
  Notify sales manager

## 6. Outputs & Handoffs
- [ ] CRM contact record created with BANT score
- [ ] First email sent (hot) or nurture sequence started (warm)
- [ ] Follow-up task created in CRM
- [ ] Sales manager notified (hot leads only)
- **Next SOP triggered:** "Qualify Prospect" (for hot leads after first response)

## 7. Exceptions & Escalation
- **Enterprise lead (500+ employees):** Skip normal scoring. Escalate directly to VP Sales.
- **Referral from existing customer:** Priority handling. Personal call within 1 hour.
- **Press/media inquiry:** Route to Marketing team, not Sales.
- **Escalation chain:** Sales Rep → Sales Manager → VP Sales
- **SLA:** First contact within 2 hours (hot), 24 hours (warm)

What makes this SOP effective

Notice how every step has an explicit expected result and a decision point where behavior branches. The lead scoring thresholds (80/50) are concrete numbers, not vague guidance. A new hire could follow this on day one without asking a single question.

Personal / Founder

Example 2: Daily Morning Review

A daily personal SOP that turns your first 15 minutes into a structured launchpad. Shows how SOPs apply to individual routines, not just team processes.

# Daily Morning Review

## 1. Metadata
- **Owner:** Founder / CEO
- **Frequency:** Daily
- **Trigger:** Start of workday (first 15 minutes)
- **Estimated Time:** 15 minutes
- **Automation Level:** Level 2

## 2. Purpose
**Why this exists:** Start each day with full clarity on priorities, outstanding commitments, and emerging opportunities.

**What happens if not done:** You react to the loudest email instead of the most important work. Days become chaotic and strategic priorities get buried.

**Success criteria:** Top 3 priorities set, inbox processed, no surprises on today's calendar.

## 3. Prerequisites
**Tools needed:**
- [ ] Sentigen dashboard - sentigen.app
- [ ] Email client - Gmail
- [ ] Calendar - Google Calendar
- [ ] Task manager - Sentigen Tasks

**Information needed:**
- [ ] Access to all communication channels

## 4. Steps

### Step 1: Check Vibe Feed
**Action:** Open Sentigen → Navigate to Vibe Feed → Review all cards from last 12 hours.
**Expected result:** Awareness of overnight activity - new tasks detected, commitments due, follow-ups needed, meeting prep available.
**If something goes wrong:** If Vibe Feed is empty, check that integrations are connected (Settings → Integrations).

### Step 2: Review overnight alerts
**Action:** Scan for any urgent alerts (red priority cards). Check for: deadline warnings, missed commitments, escalation requests.
**Decision point:**
- IF urgent alert exists → Handle immediately. Defer remaining routine.
- IF no urgent alerts → Continue to Step 3.

### Step 3: Check today's calendar
**Action:** Open Google Calendar → Review all meetings for today. For each meeting: note the time, who's attending, and whether Sentigen has generated meeting prep.
**Expected result:** Full awareness of today's time commitments. Meeting prep reviewed for important meetings.

### Step 4: Process email inbox (4-bucket system)
**Action:** Open Gmail → Process each unread email into one of 4 buckets:
- **Act** (< 2 min): Reply immediately, then archive
- **Delegate**: Forward with clear instructions, then archive
- **Schedule**: Create a task with deadline, then archive
- **Archive**: No action needed, archive immediately

**Expected result:** Inbox processed to zero. All actionable items captured as tasks or calendar events.
**Time limit:** 5 minutes maximum. If inbox is too large, process top 20 by priority.

### Step 5: Review task list
**Action:** Open Sentigen Tasks → Review all tasks due today and overdue. Check for AI-detected tasks from yesterday's communications.
**Expected result:** Complete view of all commitments and deadlines.

### Step 6: Set top 3 priorities
**Action:** From all inputs (Vibe Feed, calendar, email, tasks), identify the 3 most important outcomes for today. Write them down.
**Expected result:** Three clear, actionable priorities that would make today successful.

### Step 7: Begin priority #1
**Action:** Start working on your top priority immediately. Do not check Slack or other channels first.
**Expected result:** Deep work begins within 15 minutes of sitting down.

## 5. Decision Trees

### Meeting-heavy day
IF 4+ meetings today
THEN do abbreviated review (skip Step 4, just scan for urgent emails)
Focus on meeting prep for highest-priority meeting
Block 30-minute focus time between meetings for email processing

### Crisis mode
IF P1 alert or urgent customer issue
THEN skip morning routine entirely
Handle crisis first
Do abbreviated review during first available break

## 6. Outputs & Handoffs
- [ ] Top 3 priorities documented (visible in task manager)
- [ ] Inbox at zero
- [ ] Calendar reviewed, meeting prep consumed
- [ ] All new tasks captured from overnight communications
- **Next SOP triggered:** Specific work on Priority #1

## 7. Exceptions & Escalation
- **Monday mornings:** Extend to 25 minutes. Include weekly planning review.
- **After vacation/time off:** Extend to 45 minutes. Process accumulated items before setting priorities.
- **If you're running late:** Skip Steps 4-5. Just do Vibe Feed + Calendar + Set Priorities (5 min).

Personal SOPs are underrated

Most people think SOPs are only for teams. But your own routines are the easiest to automate and the highest leverage to document. Once this SOP is in Sentigen, the AI can surface your priorities, triage your email, and set your top 3 before you even open your laptop.

Customer Success

Example 3: Handle Customer Support Escalation

A per-event SOP with severity-based routing, SLA enforcement, and a full escalation chain. Demonstrates how to handle branching complexity in a clear, followable format.

# Handle Customer Support Escalation

## 1. Metadata
- **Owner:** Customer Success Manager
- **Frequency:** Per-event
- **Trigger:** Escalation flag in support ticket system
- **Estimated Time:** 20-60 minutes (varies by severity)
- **Automation Level:** Level 2

## 2. Purpose
**Why this exists:** Resolve escalated customer issues within SLA while maintaining customer trust and documenting learnings.

**What happens if not done:** Customer churn. Escalated customers who aren't handled within SLA have 4x higher churn rate.

**Success criteria:** Issue resolved within SLA, customer confirmed satisfied, post-mortem documented (P1 only).

## 3. Prerequisites
**Tools needed:**
- [ ] Support system (Zendesk / Intercom) - [URL]
- [ ] CRM - [URL]
- [ ] Internal communication (Slack) - #support-escalations channel
- [ ] Customer history dashboard - Sentigen CRM entity profile

**Permissions required:**
- [ ] Support ticket write access
- [ ] Authority to issue credits up to $500
- [ ] Slack access to engineering channel (P1 only)

**Information needed:**
- [ ] Customer account details and history
- [ ] Escalation ticket details
- [ ] Current system status page

## 4. Steps

### Step 1: Review escalation ticket
**Action:** Open the escalation ticket → Read full conversation history → Check customer's account: plan tier, MRR, tenure, previous escalations.
**Expected result:** Full understanding of the issue and the customer's importance.

### Step 2: Classify severity
**Action:** Apply severity matrix:
- **P1 - Critical:** Service is down or data loss. Multiple customers affected. Revenue impact.
- **P2 - High:** Feature broken, workaround exists but painful. Single customer significantly impacted.
- **P3 - Medium:** Feature degraded, reasonable workaround available. Customer inconvenienced but operational.
**Expected result:** Severity level assigned, SLA clock started.

### Step 3: Acknowledge customer
**Action:** Send acknowledgment within SLA:
- P1: Within 15 minutes
- P2: Within 1 hour
- P3: Within 4 hours

Template: "Hi [Name], I've received your escalation about [issue]. I'm personally handling this and will update you by [time]. You're a priority for us."
**Expected result:** Customer knows they're heard and has a timeline.

### Step 4: Investigate root cause
**Action:** Check system status → Review logs → Reproduce issue if possible → Check if known issue exists → Search knowledge base.
**Decision point:**
- IF known issue with existing fix → Go to Step 5
- IF unknown issue → Escalate to engineering via Slack #support-escalations with: customer name, severity, reproduction steps, impact

### Step 5: Determine resolution
**Action:** Based on investigation:
- IF fix available → Apply fix and verify
- IF fix requires engineering → Provide workaround + ETA
- IF no fix possible → Offer compensation (credit, extended trial, upgrade)
**Expected result:** Clear resolution path with timeline.

### Step 6: Execute and confirm
**Action:** Implement resolution → Test that fix works → Update customer with detailed explanation → Ask customer to confirm resolution.
**Expected result:** Customer confirms issue is resolved.

### Step 7: Document and close
**Action:** Update ticket with: root cause, resolution steps, time to resolve, customer satisfaction status → Close ticket → Update knowledge base if new issue type.
**Expected result:** Complete documentation for future reference.

### Step 8 (P1 only): Post-mortem
**Action:** Within 48 hours, write post-mortem: timeline, root cause, customer impact, what we'll do to prevent recurrence → Share in #support-escalations → Add to monthly report.

## 5. Decision Trees

### Credit authority
IF customer asks for refund or credit
  IF amount <= $500 → approve immediately, no manager approval needed
  IF amount $500-$2000 → get CS Manager approval (Slack DM, expect response within 1 hour)
  IF amount > $2000 → escalate to VP CS with full context document

### Repeat escalation
IF customer has escalated 3+ times in 90 days
  THEN flag as churn risk
  Schedule executive check-in call
  Review entire account health in CRM
  Trigger "Churn Risk Response" SOP

## 6. Outputs & Handoffs
- [ ] Ticket updated with resolution details
- [ ] Customer confirmation of resolution
- [ ] Knowledge base updated (if new issue type)
- [ ] Post-mortem document (P1 only)
- [ ] Churn risk flag (if repeat escalation)
- **Next SOP triggered:** "Churn Risk Response" (if flagged)

## 7. Exceptions & Escalation
- **VIP customer (top 10 by MRR):** Auto-escalate to P2 minimum. Notify CS Manager regardless of severity.
- **Legal threat:** Stop normal process. Escalate to VP CS + Legal immediately.
- **Data breach suspicion:** Stop normal process. Escalate to CTO + Security team immediately. Do NOT discuss details over email.
- **Escalation chain:** CS Manager → VP CS → COO
- **SLA:** P1: 1 hour to resolution. P2: 4 hours. P3: 24 hours.

Severity Routing Decision Tree

Is the service completely down or is there data loss?

Yes

P1 - Critical: 15-min acknowledgment, 1-hour resolution SLA. Page engineering immediately.

Is a core feature broken with no reasonable workaround?

Yes

P2 - High: 1-hour acknowledgment, 4-hour resolution SLA. Escalate to engineering via Slack.

P3 - Medium: 4-hour acknowledgment, 24-hour resolution SLA. Investigate and resolve within queue.

SLA is not optional

The escalation timers in this SOP are not suggestions. When you codify SLAs into your SOP, Sentigen can automatically track them, alert you before a breach, and escalate proactively. The SOP is the contract between your team and your customers.

Ready to write your own?

Use our templates as a starting point, or follow the writing guide step by step.

Browse Templates

# Process Inbound Sales Lead ## 1. Metadata - **Owner:** Sales Representative - **Frequency:** Per-event - **Trigger:** New lead form submission, email inquiry, or referral notification - **Estimated Time:** 15 minutes - **Automation Level:** Level 2 ## 2. Purpose **Why this exists:** Route and qualify inbound leads within 2 hours of receipt to maximize conversion rates. **What happens if not done:** Leads go cold. Response time is the #1 predictor of conversion - leads contacted within 5 minutes are 21x more likely to convert. **Success criteria:** Lead is scored, categorized, assigned to correct pipeline stage, and first contact is made within 2 hours. ## 3. Prerequisites **Tools needed:** - [ ] CRM (HubSpot / Salesforce) - [CRM URL] - [ ] Email client - [Email URL] - [ ] Lead scoring rubric - [Internal doc URL] **Permissions required:** - [ ] CRM write access (Contacts, Deals) - [ ] Email send permissions **Information needed:** - [ ] Lead form submission data - [ ] Current pipeline capacity ## 4. Steps ### Step 1: Check lead source **Action:** Open CRM notifications or email inbox. Identify the lead source (website form, email inquiry, referral, event). **Expected result:** You can see the lead's name, email, company, and source. **If something goes wrong:** If form data is incomplete, check the website form submission logs. ### Step 2: Review lead data **Action:** Open the lead's form submission. Note: company name, role/title, company size, stated need, how they found us. **Expected result:** Clear picture of who this person is and what they need. ### Step 3: Score the lead (BANT) **Action:** Apply BANT framework: - Budget: Can they afford the solution? (Check company size, funding stage) - Authority: Are they a decision maker? (Check title) - Need: Do they have a problem we solve? (Check stated need) - Timeline: Are they buying soon? (Check urgency signals) Score each 0-25. Total score out of 100. **Decision point:** - IF score >= 80 → Hot lead. Go to Step 4a. - IF score 50-79 → Warm lead. Go to Step 4b. - IF score < 50 → Cold lead. Go to Step 4c. ### Step 4a: Hot lead handling **Action:** Create CRM contact → Set pipeline stage to "Qualified" → Draft personalized email referencing their specific need → Schedule follow-up for next business day → Notify sales manager via Slack. **Expected result:** Contact created, email sent within 30 minutes, follow-up scheduled. ### Step 4b: Warm lead handling **Action:** Create CRM contact → Set pipeline stage to "Nurture" → Add to automated nurture email sequence → Schedule manual check-in for 1 week. **Expected result:** Contact in nurture sequence, check-in scheduled. ### Step 4c: Cold lead handling **Action:** Create CRM contact → Set pipeline stage to "Newsletter" → Add to monthly newsletter list → No immediate follow-up. **Expected result:** Contact created, added to newsletter. ### Step 5: Log activity **Action:** In CRM, log: lead source, BANT score, pipeline stage assigned, first action taken, notes on conversation. **Expected result:** Complete activity log visible to entire sales team. ## 5. Decision Trees ### Duplicate lead check IF lead email already exists in CRM THEN check last activity date IF last activity > 90 days → treat as new lead, update existing record IF last activity < 90 days → notify original owner, do not re-assign ### Competitor inquiry IF lead is from a known competitor (check against competitor list) THEN do NOT add to pipeline Log as "Competitor Inquiry" in CRM Notify sales manager ## 6. Outputs & Handoffs - [ ] CRM contact record created with BANT score - [ ] First email sent (hot) or nurture sequence started (warm) - [ ] Follow-up task created in CRM - [ ] Sales manager notified (hot leads only) - **Next SOP triggered:** "Qualify Prospect" (for hot leads after first response) ## 7. Exceptions & Escalation - **Enterprise lead (500+ employees):** Skip normal scoring. Escalate directly to VP Sales. - **Referral from existing customer:** Priority handling. Personal call within 1 hour. - **Press/media inquiry:** Route to Marketing team, not Sales. - **Escalation chain:** Sales Rep → Sales Manager → VP Sales - **SLA:** First contact within 2 hours (hot), 24 hours (warm)

# Daily Morning Review ## 1. Metadata - **Owner:** Founder / CEO - **Frequency:** Daily - **Trigger:** Start of workday (first 15 minutes) - **Estimated Time:** 15 minutes - **Automation Level:** Level 2 ## 2. Purpose **Why this exists:** Start each day with full clarity on priorities, outstanding commitments, and emerging opportunities. **What happens if not done:** You react to the loudest email instead of the most important work. Days become chaotic and strategic priorities get buried. **Success criteria:** Top 3 priorities set, inbox processed, no surprises on today's calendar. ## 3. Prerequisites **Tools needed:** - [ ] Sentigen dashboard - sentigen.app - [ ] Email client - Gmail - [ ] Calendar - Google Calendar - [ ] Task manager - Sentigen Tasks **Information needed:** - [ ] Access to all communication channels ## 4. Steps ### Step 1: Check Vibe Feed **Action:** Open Sentigen → Navigate to Vibe Feed → Review all cards from last 12 hours. **Expected result:** Awareness of overnight activity - new tasks detected, commitments due, follow-ups needed, meeting prep available. **If something goes wrong:** If Vibe Feed is empty, check that integrations are connected (Settings → Integrations). ### Step 2: Review overnight alerts **Action:** Scan for any urgent alerts (red priority cards). Check for: deadline warnings, missed commitments, escalation requests. **Decision point:** - IF urgent alert exists → Handle immediately. Defer remaining routine. - IF no urgent alerts → Continue to Step 3. ### Step 3: Check today's calendar **Action:** Open Google Calendar → Review all meetings for today. For each meeting: note the time, who's attending, and whether Sentigen has generated meeting prep. **Expected result:** Full awareness of today's time commitments. Meeting prep reviewed for important meetings. ### Step 4: Process email inbox (4-bucket system) **Action:** Open Gmail → Process each unread email into one of 4 buckets: - **Act** (< 2 min): Reply immediately, then archive - **Delegate**: Forward with clear instructions, then archive - **Schedule**: Create a task with deadline, then archive - **Archive**: No action needed, archive immediately **Expected result:** Inbox processed to zero. All actionable items captured as tasks or calendar events. **Time limit:** 5 minutes maximum. If inbox is too large, process top 20 by priority. ### Step 5: Review task list **Action:** Open Sentigen Tasks → Review all tasks due today and overdue. Check for AI-detected tasks from yesterday's communications. **Expected result:** Complete view of all commitments and deadlines. ### Step 6: Set top 3 priorities **Action:** From all inputs (Vibe Feed, calendar, email, tasks), identify the 3 most important outcomes for today. Write them down. **Expected result:** Three clear, actionable priorities that would make today successful. ### Step 7: Begin priority #1 **Action:** Start working on your top priority immediately. Do not check Slack or other channels first. **Expected result:** Deep work begins within 15 minutes of sitting down. ## 5. Decision Trees ### Meeting-heavy day IF 4+ meetings today THEN do abbreviated review (skip Step 4, just scan for urgent emails) Focus on meeting prep for highest-priority meeting Block 30-minute focus time between meetings for email processing ### Crisis mode IF P1 alert or urgent customer issue THEN skip morning routine entirely Handle crisis first Do abbreviated review during first available break ## 6. Outputs & Handoffs - [ ] Top 3 priorities documented (visible in task manager) - [ ] Inbox at zero - [ ] Calendar reviewed, meeting prep consumed - [ ] All new tasks captured from overnight communications - **Next SOP triggered:** Specific work on Priority #1 ## 7. Exceptions & Escalation - **Monday mornings:** Extend to 25 minutes. Include weekly planning review. - **After vacation/time off:** Extend to 45 minutes. Process accumulated items before setting priorities. - **If you're running late:** Skip Steps 4-5. Just do Vibe Feed + Calendar + Set Priorities (5 min).

# Handle Customer Support Escalation ## 1. Metadata - **Owner:** Customer Success Manager - **Frequency:** Per-event - **Trigger:** Escalation flag in support ticket system - **Estimated Time:** 20-60 minutes (varies by severity) - **Automation Level:** Level 2 ## 2. Purpose **Why this exists:** Resolve escalated customer issues within SLA while maintaining customer trust and documenting learnings. **What happens if not done:** Customer churn. Escalated customers who aren't handled within SLA have 4x higher churn rate. **Success criteria:** Issue resolved within SLA, customer confirmed satisfied, post-mortem documented (P1 only). ## 3. Prerequisites **Tools needed:** - [ ] Support system (Zendesk / Intercom) - [URL] - [ ] CRM - [URL] - [ ] Internal communication (Slack) - #support-escalations channel - [ ] Customer history dashboard - Sentigen CRM entity profile **Permissions required:** - [ ] Support ticket write access - [ ] Authority to issue credits up to $500 - [ ] Slack access to engineering channel (P1 only) **Information needed:** - [ ] Customer account details and history - [ ] Escalation ticket details - [ ] Current system status page ## 4. Steps ### Step 1: Review escalation ticket **Action:** Open the escalation ticket → Read full conversation history → Check customer's account: plan tier, MRR, tenure, previous escalations. **Expected result:** Full understanding of the issue and the customer's importance. ### Step 2: Classify severity **Action:** Apply severity matrix: - **P1 - Critical:** Service is down or data loss. Multiple customers affected. Revenue impact. - **P2 - High:** Feature broken, workaround exists but painful. Single customer significantly impacted. - **P3 - Medium:** Feature degraded, reasonable workaround available. Customer inconvenienced but operational. **Expected result:** Severity level assigned, SLA clock started. ### Step 3: Acknowledge customer **Action:** Send acknowledgment within SLA: - P1: Within 15 minutes - P2: Within 1 hour - P3: Within 4 hours Template: "Hi [Name], I've received your escalation about [issue]. I'm personally handling this and will update you by [time]. You're a priority for us." **Expected result:** Customer knows they're heard and has a timeline. ### Step 4: Investigate root cause **Action:** Check system status → Review logs → Reproduce issue if possible → Check if known issue exists → Search knowledge base. **Decision point:** - IF known issue with existing fix → Go to Step 5 - IF unknown issue → Escalate to engineering via Slack #support-escalations with: customer name, severity, reproduction steps, impact ### Step 5: Determine resolution **Action:** Based on investigation: - IF fix available → Apply fix and verify - IF fix requires engineering → Provide workaround + ETA - IF no fix possible → Offer compensation (credit, extended trial, upgrade) **Expected result:** Clear resolution path with timeline. ### Step 6: Execute and confirm **Action:** Implement resolution → Test that fix works → Update customer with detailed explanation → Ask customer to confirm resolution. **Expected result:** Customer confirms issue is resolved. ### Step 7: Document and close **Action:** Update ticket with: root cause, resolution steps, time to resolve, customer satisfaction status → Close ticket → Update knowledge base if new issue type. **Expected result:** Complete documentation for future reference. ### Step 8 (P1 only): Post-mortem **Action:** Within 48 hours, write post-mortem: timeline, root cause, customer impact, what we'll do to prevent recurrence → Share in #support-escalations → Add to monthly report. ## 5. Decision Trees ### Credit authority IF customer asks for refund or credit IF amount <= $500 → approve immediately, no manager approval needed IF amount $500-$2000 → get CS Manager approval (Slack DM, expect response within 1 hour) IF amount > $2000 → escalate to VP CS with full context document ### Repeat escalation IF customer has escalated 3+ times in 90 days THEN flag as churn risk Schedule executive check-in call Review entire account health in CRM Trigger "Churn Risk Response" SOP ## 6. Outputs & Handoffs - [ ] Ticket updated with resolution details - [ ] Customer confirmation of resolution - [ ] Knowledge base updated (if new issue type) - [ ] Post-mortem document (P1 only) - [ ] Churn risk flag (if repeat escalation) - **Next SOP triggered:** "Churn Risk Response" (if flagged) ## 7. Exceptions & Escalation - **VIP customer (top 10 by MRR):** Auto-escalate to P2 minimum. Notify CS Manager regardless of severity. - **Legal threat:** Stop normal process. Escalate to VP CS + Legal immediately. - **Data breach suspicion:** Stop normal process. Escalate to CTO + Security team immediately. Do NOT discuss details over email. - **Escalation chain:** CS Manager → VP CS → COO - **SLA:** P1: 1 hour to resolution. P2: 4 hours. P3: 24 hours.