Churn Risk Score¶

Definition¶

Churn Risk Score is a predictive metric that estimates the likelihood of a customer canceling or downgrading within a given period. It helps identify at-risk accounts for proactive retention efforts.

Description¶

Churn Risk Score is a predictive metric that combines behavioral, financial, and support signals to assess how likely a customer is to churn — enabling proactive retention strategies and health-based account workflows.

The relevance and interpretation of this metric shift depending on the model or product:

In SaaS, it often includes usage decline, ticket sentiment, missed milestones
In eCommerce, it reflects purchase gaps or abandoned cart behavior
In enterprise, it may involve renewal silence or stakeholder disengagement

A rising risk score signals potential churn triggers. A declining score reflects successful recovery or engagement. Segment by plan tier, vertical, or behavior type to identify which accounts need attention — and what’s driving improvement.

Churn Risk Score informs:

Strategic decisions, like forecasting revenue risk or staffing CS teams
Tactical actions, such as triggering lifecycle campaigns or CSM alerts
Operational improvements, including flagging UX issues or friction in onboarding
Cross-functional alignment, by syncing product, CS, and lifecycle teams on early warning signs and recovery paths

Key Drivers¶

These are the main factors that directly impact the metric. Understanding these lets you know what levers you can pull to improve the outcome

Product Usage Patterns: Declining logins, fewer active users, or skipped milestones are leading indicators of churn risk. Inactivity ≠ satisfaction.
Support and Satisfaction Signals: Unresolved tickets, low NPS, or high complaint volume increase risk — even if usage looks steady.
Change in Internal Champions or Behavior: When power users leave or usage shifts to new (unengaged) roles, churn risk spikes.

Improvement Tactics & Quick Wins¶

Actionable ideas to optimize this KPI, from fast, low-effort wins to strategic initiatives that drive measurable impact.

If churn risk scores are rising, trigger early interventions (check-ins, success nudges, help content) based on usage dips.
Add real-time alerts in your CRM or CS tools for high-risk signals, such as login drop-offs or failed integrations.
Run a test offering free strategic consultations to at-risk accounts, and monitor retention uplift.
Refine the scoring model to weigh key moments (e.g., Day 30 logins, invite count, milestone completion) more heavily.
Partner with CS to build playbooks tied to score thresholds, ensuring consistent, proactive follow-up.

Required Datapoints to calculate the metric
- Product Usage Frequency & Depth
- Login/Activity Trends
- Support Interactions & CSAT
- Contract or Billing Status
- Account Engagement (e.g. check-ins, QBRs)
- Custom Risk Triggers (e.g., key user churned)
Example to show how the metric is derived

For Account A:
- Usage down 60%
- No logins in 14 days
- Missed 2 check-ins
- No product adoption in onboarding Churn Risk Score = 87/100

Formula¶

Formula

\[ \mathrm{Churn\ Risk\ Score} = \left( \sum_{i=1}^{n} w_i \cdot x_i \right) \quad \text{or} \quad \mathrm{Probability\ of\ Churn} = \left( \frac{\mathrm{Number\ of\ Churned\ Accounts}}{\mathrm{Total\ Accounts}} \right) \times 100 \]

Data Model Definition¶

How this KPI is structured in Cube.js, including its key measures, dimensions, and calculation logic for consistent reporting.

CustomerChurnRisk

cube('CustomerChurnRisk', {
  sql: `SELECT * FROM customer_churn_risk`,

  joins: {
    ProductUsage: {
      relationship: 'belongsTo',
      sql: `${CUBE}.customer_id = ${ProductUsage}.customer_id`
    },
    SupportInteractions: {
      relationship: 'belongsTo',
      sql: `${CUBE}.customer_id = ${SupportInteractions}.customer_id`
    },
    BillingStatus: {
      relationship: 'belongsTo',
      sql: `${CUBE}.customer_id = ${BillingStatus}.customer_id`
    },
    AccountEngagement: {
      relationship: 'belongsTo',
      sql: `${CUBE}.customer_id = ${AccountEngagement}.customer_id`
    }
  },

  measures: {
    churnRiskScore: {
      sql: `churn_risk_score`,
      type: 'number',
      title: 'Churn Risk Score',
      description: 'Predictive score estimating the likelihood of a customer canceling or downgrading.'
    }
  },

  dimensions: {
    customerId: {
      sql: `customer_id`,
      type: 'string',
      primaryKey: true,
      title: 'Customer ID',
      description: 'Unique identifier for each customer.'
    },
    productUsageFrequency: {
      sql: `product_usage_frequency`,
      type: 'number',
      title: 'Product Usage Frequency',
      description: 'Frequency of product usage by the customer.'
    },
    productUsageDepth: {
      sql: `product_usage_depth`,
      type: 'number',
      title: 'Product Usage Depth',
      description: 'Depth of product usage by the customer.'
    },
    loginActivityTrend: {
      sql: `login_activity_trend`,
      type: 'number',
      title: 'Login/Activity Trend',
      description: 'Trend of customer login and activity over time.'
    },
    supportInteractionsCount: {
      sql: `support_interactions_count`,
      type: 'number',
      title: 'Support Interactions Count',
      description: 'Number of support interactions by the customer.'
    },
    csatScore: {
      sql: `csat_score`,
      type: 'number',
      title: 'CSAT Score',
      description: 'Customer satisfaction score from support interactions.'
    },
    contractStatus: {
      sql: `contract_status`,
      type: 'string',
      title: 'Contract Status',
      description: 'Current contract status of the customer.'
    },
    billingStatus: {
      sql: `billing_status`,
      type: 'string',
      title: 'Billing Status',
      description: 'Current billing status of the customer.'
    },
    accountEngagementLevel: {
      sql: `account_engagement_level`,
      type: 'number',
      title: 'Account Engagement Level',
      description: 'Level of engagement with the account, such as check-ins and QBRs.'
    },
    customRiskTriggers: {
      sql: `custom_risk_triggers`,
      type: 'string',
      title: 'Custom Risk Triggers',
      description: 'Custom triggers indicating potential risk, such as key user churn.'
    },
    createdAt: {
      sql: `created_at`,
      type: 'time',
      title: 'Created At',
      description: 'Timestamp when the record was created.'
    }
  }
});

Note: This is a reference implementation and should be used as a starting point. You’ll need to adapt it to match your own data model and schema

Positive & Negative Influences¶

Negative influences

Factors that drive the metric in an undesirable direction, often signaling risk or decline.
- Declining Logins: A decrease in the frequency of logins indicates reduced engagement, which is a strong predictor of increased churn risk.
- Unresolved Support Tickets: High volume of unresolved support tickets suggests dissatisfaction, directly increasing the churn risk score.
- Low Net Promoter Score (NPS): A low NPS reflects customer dissatisfaction and correlates with a higher likelihood of churn.
- High Complaint Volume: An increase in complaint volume is a clear indicator of customer dissatisfaction, leading to a higher churn risk score.
- Loss of Internal Champions: When key users or advocates leave, the risk of churn increases as the remaining users may not be as engaged.
Positive influences

Factors that push the metric in a favorable direction, supporting growth or improvement.
- Increased Product Usage: Higher frequency of product usage indicates strong engagement, which reduces the churn risk score.
- Resolved Support Tickets: Timely resolution of support tickets enhances customer satisfaction, thereby lowering the churn risk score.
- High Net Promoter Score (NPS): A high NPS indicates customer satisfaction and loyalty, which decreases the likelihood of churn.
- Low Complaint Volume: A decrease in complaint volume suggests improved customer satisfaction, reducing the churn risk score.
- Engagement of New Internal Champions: When new users become engaged and take on champion roles, it decreases the churn risk by ensuring continued product advocacy.

Involved Roles & Activities¶

Involved Roles

These roles are typically responsible for implementing or monitoring this KPI:

Customer Success
Customer Lifecycle Management
Product Management (PM)
Revenue Operations
Activities

Common initiatives or actions associated with this KPI:

Churn Reduction
Renewal Forecasting
CS Ops
Predictive Analytics

Funnel Stage & Type¶

AAARRR Funnel Stage

This KPI is associated with the following stages in the AAARRR (Pirate Metrics) funnel:

Retention
Type

This KPI is classified as a Lagging Indicator. It reflects the results of past actions or behaviors and is used to validate performance or assess the impact of previous strategies.

Supporting Leading & Lagging Metrics¶

Leading

These leading indicators influence this KPI and act as early signals that forecast future changes in this KPI.
- Customer Health Score: Customer Health Score is a composite early-warning signal that integrates product usage, engagement, and satisfaction. A declining health score often precedes a rise in Churn Risk Score, allowing teams to intervene before risk escalates.
- Product Qualified Leads: A drop in Product Qualified Leads (PQLs) signals reduced product engagement and value realization, typically resulting in higher Churn Risk Scores as fewer users are demonstrating behaviors that indicate satisfaction and commitment.
- Activation Rate: Activation Rate reflects how many users reach initial value milestones. Low or declining activation rates indicate onboarding friction, which increases the likelihood of future churn and raises the Churn Risk Score.
- Customer Loyalty: Customer Loyalty measures the likelihood of repeat engagement and purchase. Lower loyalty is a predictive signal of potential churn, directly influencing the Churn Risk Score before actual cancellations occur.
- Stickiness Ratio: Stickiness Ratio (DAU/MAU) reveals habitual product use. A falling stickiness ratio signals disengagement, which is often an early indicator that feeds into and elevates the Churn Risk Score.
Lagging

These lagging indicators confirm, quantify, or amplify this KPI and help explain the broader business impact on this KPI after the fact.
- Customer Downgrade Rate: Rising Customer Downgrade Rate confirms that users are not finding enough value, validating elevated Churn Risk Scores and quantifying actual account contraction that typically precedes churn.
- Customer Churn Rate: Customer Churn Rate provides a hard outcome measure tied to the risks identified by Churn Risk Score, confirming the predictive power and calibration of the risk model.
- Net Revenue Churn: Net Revenue Churn quantifies the real revenue impact of churn and downgrades, adding financial context and confirming the wider business consequences of high Churn Risk Scores.
- Contract Renewal Rate: Contract Renewal Rate verifies the predictive accuracy of Churn Risk Score by showing the proportion of at-risk accounts that actually renew or churn, closing the loop on risk assessment.
- Customer Feedback Retention Score: This metric tracks retention among those who provide feedback. If at-risk accounts (high Churn Risk Score) give negative feedback and then leave, this score validates and helps calibrate the predictive model.