SaaS Churn Prevention

Use behavioral analytics to identify churn signals early and take action to retain customers.

Understanding SaaS Churn

Churn Types

Type	Definition	Trackable?
Voluntary	Customer cancels	Yes
Involuntary	Payment fails	Yes
Quiet	Stops using, keeps paying	Yes - critical!
Downgrade	Moves to lower plan	Yes

Churn Metrics

Metric	Formula	Illustrative range
Monthly Churn Rate	Lost ÷ Start of Month	<5%
Annual Churn	Lost ÷ Annual Customers	<10%
Net Revenue Retention	(MRR + Expansion - Churn) ÷ Starting MRR	>100%
Logo Churn	Lost accounts ÷ Total accounts	<3%

Identifying Churn Signals

Behavioral Signals

Track usage patterns that predict churn:

Signal	Risk Level	Days Before Churn
Zero logins (7 days)	High	14-30
Decreased feature use	Medium	30-60
Support ticket spike	Medium	30-45
Admin change	Watch	45-60
Integration removed	High	14-21

Track Activity Decline

// Track daily active usage
zenovay('track','user_session', {
  user_id: 'user_123',
  account_id: 'acc_456',
  session_duration: 1200,
  features_used: ['dashboard', 'reports', 'export'],
  actions_count: 45
});

// Track feature removal
zenovay('track','feature_disabled', {
  account_id: 'acc_456',
  feature: 'api_integration',
  connected_since_days: 180
});

Create Health Score

// Calculate and track health score
function updateAccountHealth(accountId) {
  const factors = {
    login_frequency: calculateLoginScore(),      // 0-25
    feature_breadth: calculateFeatureScore(),    // 0-25
    user_growth: calculateUserGrowthScore(),     // 0-20
    support_sentiment: calculateSupportScore(),  // 0-15
    billing_health: calculateBillingScore()      // 0-15
  };

  const totalScore = Object.values(factors).reduce((a, b) => a + b, 0);

  zenovay('track','health_score_updated', {
    account_id: accountId,
    score: totalScore,
    factors: factors,
    risk_level: totalScore < 40 ? 'high' : totalScore < 70 ? 'medium' : 'low'
  });
}

Building a Churn Dashboard

Risk Segments

Create segments based on health scores:

Segment	Health Score	% of Base	Action
Healthy	80-100	45%	Upsell
Stable	60-79	30%	Monitor
At Risk	40-59	15%	Intervene
Critical	0-39	10%	Urgent

Track Segment Movement

// Track when account changes risk level
zenovay('track','risk_level_changed', {
  account_id: 'acc_456',
  previous_level: 'stable',
  new_level: 'at_risk',
  trigger: 'login_frequency_dropped',
  days_at_previous: 90
});

Set Up Alerts

Alert when:

• Account enters "At Risk" segment
• Health score drops >20 points in 7 days
• Zero activity for key accounts
• Multiple users from same account inactive

Usage Pattern Analysis

Login Frequency Tracking

// Track login patterns
zenovay('track','user_login', {
  user_id: 'user_123',
  account_id: 'acc_456',
  days_since_last_login: 3,
  login_streak: 5,
  device: 'desktop'
});

Feature Usage Depth

Track breadth of feature usage:

Usage Level	Features Used	Churn Risk
Power User	80%+ features	Very Low
Regular	50-79%	Low
Basic	25-49%	Medium
Minimal	<25%	High

// Track feature engagement
zenovay('track','feature_engagement', {
  account_id: 'acc_456',
  features_available: 20,
  features_used_30d: 8,
  engagement_score: 40,
  trending: 'declining'
});

Key Feature Stickiness

Identify features that correlate with retention:

Feature	Users Who Use	12-mo Retention
Team collaboration	45%	92%
API integration	30%	88%
Custom reports	55%	85%
Basic dashboard only	70%	58%

Track adoption of sticky features:

zenovay('track','sticky_feature_adopted', {
  account_id: 'acc_456',
  feature: 'team_collaboration',
  users_using: 5,
  adoption_date: new Date().toISOString()
});

Cohort Churn Analysis

Signup Cohorts

Track churn by signup month:

Cohort	M1	M3	M6	M12
Jan	5%	12%	18%	25%
Feb	4%	10%	15%	-
Mar	6%	14%	-	-

By Acquisition Channel

Channel	12-mo Churn	Notes
Referral	15%	Best retention
Organic	22%	Good fit
Paid Ads	35%	Lower quality
Affiliate	40%	Price-driven

By Plan Type

Plan	Annual Churn	MRR Churn
Monthly	28%	25%
Annual	8%	10%
Enterprise	5%	8%

Pre-Churn Indicators

Cancellation Flow Tracking

// User opens cancellation page
zenovay('track','cancellation_page_viewed', {
  account_id: 'acc_456',
  current_plan: 'pro',
  months_as_customer: 8,
  mrr: 99
});

// User provides reason
zenovay('track','cancellation_reason', {
  account_id: 'acc_456',
  reason: 'too_expensive',
  feedback: 'Need more flexible pricing'
});

// Cancellation completed
zenovay('track','subscription_cancelled', {
  account_id: 'acc_456',
  reason: 'too_expensive',
  lifetime_value: 792,
  months_as_customer: 8,
  cancel_date: new Date().toISOString(),
  effective_date: '2025-02-15'
});

Save Offer Tracking

// Offer shown during cancellation
zenovay('track','save_offer_shown', {
  account_id: 'acc_456',
  offer_type: 'discount_30',
  offer_value: 30
});

// Offer accepted
zenovay('track','save_offer_accepted', {
  account_id: 'acc_456',
  offer_type: 'discount_30',
  new_mrr: 69,
  commitment_months: 3
});

Proactive Intervention

Triggered Actions

Set up automated interventions:

Trigger	Action	Track
7 days no login	Email check-in	intervention_email_sent
Feature usage down 50%	In-app tips	feature_tips_shown
Support ticket negative	CSM outreach	csm_assigned
Health score <40	Personal call	intervention_call

// Track intervention
zenovay('track','intervention_triggered', {
  account_id: 'acc_456',
  trigger: 'health_score_drop',
  intervention_type: 'csm_outreach',
  priority: 'high'
});

// Track intervention outcome
zenovay('track','intervention_outcome', {
  account_id: 'acc_456',
  intervention_type: 'csm_outreach',
  outcome: 'saved',
  new_health_score: 65,
  notes: 'Resolved integration issue'
});

Win-Back Campaigns

Track churned customer recovery:

// Win-back offer sent
zenovay('track','winback_campaign_sent', {
  account_id: 'acc_456',
  days_since_churn: 30,
  offer: 'first_month_free'
});

// Customer reactivated
zenovay('track','customer_reactivated', {
  account_id: 'acc_456',
  days_churned: 45,
  reactivation_offer: 'first_month_free',
  new_plan: 'pro'
});

Churn Analysis Reports

Monthly Churn Report

Include:

• Churned accounts list
• Churn reasons breakdown
• Revenue impact
• Health score before churn
• Early warning accuracy

Cohort Retention Report

Track long-term retention:

• Month 1, 3, 6, 12 retention
• By channel, plan, company size
• Trend analysis

Intervention Effectiveness

Intervention	Triggered	Saved	Save Rate
Email series	150	45	30%
CSM outreach	50	25	50%
Discount offer	80	32	40%
Training session	30	18	60%

Predictive Churn Model

Input Signals

Feed these metrics to predict churn:

• Login frequency trend
• Feature usage breadth
• Support ticket volume/sentiment
• Billing issues
• User count trend
• Admin activity

Risk Scoring

// Daily risk score calculation
zenovay('track','churn_prediction', {
  account_id: 'acc_456',
  churn_probability: 0.35,
  confidence: 0.82,
  top_factors: [
    'login_decline',
    'feature_usage_narrow',
    'support_negative'
  ],
  predicted_churn_date: '2025-03-15'
});

Model Performance

Track prediction accuracy:

• True positives (predicted and churned)
• False positives (predicted but stayed)
• Precision and recall
• Lead time (how early we predicted)

Best Practices

Early Detection

1. Monitor continuously

   • Daily health score updates
   • Weekly trend analysis
   • Monthly cohort review

2. Multiple signals

   • Don't rely on one metric
   • Combine usage + support + billing

3. Act quickly

   • Intervene at first warning
   • Don't wait for confirmation

Intervention Strategy

1. Segment approach

   • High MRR = personal touch
   • Low MRR = automated

2. Understand why

   • Root cause, not symptoms
   • Product issue vs fit issue

3. Measure effectiveness

   • Track intervention outcomes
   • Optimize based on data

Common Mistakes

1. Focusing only on cancellations

   • Quiet churn is bigger
   • Track usage, not just subscription

2. Same treatment for all

   • Different segments need different approaches
   • Personalize interventions

3. Reacting too late

   • When they cancel, it's often too late
   • Early warning is key

Next Steps

• Feature Adoption Metrics
• AI Insights for Anomalies
• Revenue Attribution

Use this workflow in Zenovay

Use user segments and goals in Open the dashboard to compare declining use with renewal outcomes. Treat an at-risk label as a hypothesis to test.