Dashboards
Churn model dashboard
Dashboards
Churn model dashboard
Documentation for Churn model dashboard
This dashboard presents performance metrics for historical churn periods, integrating model scores and actual results from past periods. It provides insights into how the model’s predictions align with real outcomes, helping to evaluate and optimize future periods.
Metric Explanations (Top of the Dashboard):
Churn Rate
- Explanation: The churn rate represents the percentage of total subscribers who have left (churned) during the selected period.
- Current Example: In this case, the churn rate is 2.11%, meaning 2.11% of the total subscribers have churned during the selected time frame.
Churn Rate
- Explanation: The churn rate represents the percentage of total subscribers who have left (churned) during the selected period.
- Current Example: In this case, the churn rate is 2.11%, meaning 2.11% of the total subscribers have churned during the selected time frame.
Total Churners
- Explanation: This metric shows the total number of subscribers who have churned during the selected period. It represents the absolute number of customers lost during this time.
- Current Example: Here, 1,003 subscribers have churned during the selected period.
Total Subscriptions
- Explanation: The total number of subscriptions reflects the total base of subscribers for the selected time period. This is the pool from which churners are measured.
- Current Example: There are 47,549 total subscriptions during this period.
Predicted Positives
- Explanation: This is a predicted metric that estimates how many of the total subscriptions are likely to churn, based on the model’s predictions. It gives users an idea of the possible future churn volume.
- Current Example: The model predicts that 1,080.51 subscribers are likely to churn.
Graph Explanations:
Base Split per Score Group
- Graph Explanation: This bar chart shows how the total subscription base is distributed across different score groups. Score groups rank subscribers based on their likelihood to churn, with each bar representing the number of subscribers in that group.
- Current Example: Each group contains around 4,754 to 4,756 subscribers, showing an even distribution of the total subscription base across score groups.
Base Split per Score Group
- Graph Explanation: This bar chart shows how the total subscription base is distributed across different score groups. Score groups rank subscribers based on their likelihood to churn, with each bar representing the number of subscribers in that group.
- Current Example: Each group contains around 4,754 to 4,756 subscribers, showing an even distribution of the total subscription base across score groups.
Churners Split per Score Group
- Graph Explanation: This chart shows how churners are distributed across different score groups. Each bar represents the number of churners in a particular score group, indicating which groups contribute the most to churn.
- Current Example: The top-ranked group (group 10) contains 281 churners, while lower-ranked groups contain fewer churners, with the lowest-ranked group (group 100) containing 33 churners.
Churn Rate per Score Group
- Graph Explanation: This bar chart shows the churn rate for each score group. It highlights the percentage of subscribers who churned within each group and compares it to the average churn rate across all groups.
- Current Example: The highest churn rate is 3.68% in score group 10, while the churn rate decreases in lower-ranked groups, with the lowest rate being 0.69% in group 100. The average churn rate is shown as a reference.
Interactive Features Explanations:
EntityID Selection (Time Period)
- Explanation: The EntityID selection dropdown allows users to select a specific time period for analysis. Once selected, the dashboard updates to show churn data for that particular period.
- Interactive Action: Users can select a different time period (e.g., Q4 2023, Q1 2024) to view churn rates, total churners, and other metrics for that specific period. The Record Count column shows the total number of subscriptions for each period, allowing for comparisons across time periods.
EntityID Selection (Time Period)
- Explanation: The EntityID selection dropdown allows users to select a specific time period for analysis. Once selected, the dashboard updates to show churn data for that particular period.
- Interactive Action: Users can select a different time period (e.g., Q4 2023, Q1 2024) to view churn rates, total churners, and other metrics for that specific period. The Record Count column shows the total number of subscriptions for each period, allowing for comparisons across time periods.
Selection Optimization Slider
- Explanation: This slider allows users to adjust the percentage of subscribers included in the analysis. By moving the slider, users can see how different levels of optimization (targeted actions) impact churn rates and other metrics.
- Interactive Action: Adjusting the slider dynamically updates the dashboard to reflect the new selection, providing a clearer understanding of the impact of focusing on specific score groups.
Key Takeaways:
- Churn Distribution Across Score Groups: The dashboard provides a clear visualization of how churn is distributed across different score groups, helping users identify which segments of the customer base are more likely to churn.
- Score Group Insights: By analyzing the churn rate per score group, users can easily spot the most vulnerable groups (e.g., the top 10% most likely to churn) and prioritize actions or campaigns to retain these customers.
- Time Period Comparison: Users can select different time periods using the EntityID Selection feature and compare churn rates and total churners across various time frames, enabling better planning and decision-making for future retention strategies.
Churn model second tab: Churn characteristics
Metric Explanations (Top of the Dashboard):
Churn Rate
- Explanation: The churn rate represents the percentage of total customers or subscribers who have left (churned) during the selected period. It is a key indicator of customer retention performance.
Churn Rate
- Explanation: The churn rate represents the percentage of total customers or subscribers who have left (churned) during the selected period. It is a key indicator of customer retention performance.
Total Churners
- Explanation: The total number of churners indicates how many customers or subscribers have left during the selected period. This is the absolute count of churners based on the available data.
Total Subscriptions
- Explanation: The total number of subscriptions or customer base from which the churn rate is calculated. It shows the overall customer population during the period.
Graph Explanations (Applies to All Types of Data):
Total Churners vs. Churn Rate Graphs
- Explanation: These bar charts show two key metrics:
- Total Churners (displayed as bars): The number of churners for a specific group or bin, representing how many customers churned from each group.
- Churn Rate (displayed as a line): The percentage of customers who churned relative to the total number in that group or bin.
Total Churners vs. Churn Rate Graphs
- Explanation: These bar charts show two key metrics:
- Total Churners (displayed as bars): The number of churners for a specific group or bin, representing how many customers churned from each group.
- Churn Rate (displayed as a line): The percentage of customers who churned relative to the total number in that group or bin.
Example of Graph Types: Amount Bins
- Explanation: If the data includes subscription amounts, this graph might display how churn is distributed across different payment levels. Users can observe if churn is higher in lower or higher-paying customer segments.
Example of Graph Types: Duration Bins
- Explanation: If the data tracks how long customers have been with the company, this graph might show the relationship between customer tenure and churn. It can reveal whether newer or long-term customers are more likely to churn.
Example of Graph Types: Demographic Bins
- Explanation: If customer data includes demographics like gender, age, or other categories, the graphs will show churn distribution and rates across these attributes. This helps in targeting specific demographics for retention efforts.
Example of Graph Types: Campaign Bins
- Explanation: If the data includes information about recruitment or marketing campaigns, the graphs will display churn distribution for customers who joined via different campaigns. This can help assess campaign effectiveness in terms of customer retention.
Interactive Sliders Explanation (Generic):
Selection Slider
- Explanation: This slider lets users adjust the selection of data included in the analysis. By narrowing or expanding the selection, users can focus on different segments of the customer base and see how churn is affected.
Selection Slider
- Explanation: This slider lets users adjust the selection of data included in the analysis. By narrowing or expanding the selection, users can focus on different segments of the customer base and see how churn is affected.
Attribute Bins (Amount, Duration, Age, etc.)
- Explanation: These sliders allow users to adjust the ranges for each specific attribute (e.g., amount paid, subscription duration, or age). Adjusting the sliders dynamically updates the corresponding graphs, letting users explore different groupings of data and see how churn varies across different segments.
Flexible Bin Names
- Depending on the customer’s data, the bins can represent different attributes:
- Amount Bins could represent different ranges of customer spending.
- Duration Bins could represent how long customers have been subscribed.
- Demographic Bins could represent different age groups, genders, or other demographic factors.
- Campaign Bins could represent different recruitment or marketing campaigns.
EntityID Selection (Time Periods)
- Explanation: Users can select different time periods (e.g., quarterly or yearly) for analysis using the EntityID Selection dropdown. This updates the dashboard to reflect churn data for the selected period, allowing users to compare churn rates and totals over different time frames.
Key Takeaways:
- Customizable Insights: This tab allows users to customize their churn analysis based on available data attributes, making it versatile across different types of businesses. Whether focusing on demographics, subscription levels, or campaign performance, users can easily adjust the dashboard to suit their specific needs.
- Targeted Retention Strategies: By visualizing churn data across different segments of the customer base, users can identify at-risk groups and tailor retention strategies accordingly. The dashboard provides flexibility in exploring various customer attributes, helping businesses focus their efforts where churn is highest.
- Time-Based Comparisons: With the ability to select different time periods, users can track how churn trends change over time. This enables better planning and understanding of customer behavior, seasonality, and the impact of retention efforts across multiple periods.