> For the complete documentation index, see [llms.txt](https://doc.batch.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://doc.batch.com/guides-and-best-practices/batch-ai/how-to-identify-and-retain-customers-at-risk-of-cancelling-their-subscription.md). # How to Identify and Retain Customers at Risk of Cancelling Their Subscription ## What Is the Subscription Churn Score? Batch AI Predict's **Subscription Churn** score estimates, for each subscriber in your database, the probability that they will cancel their subscription before it reaches its end date. The output is a float between 0 and 1: the closer to 1, the higher the predicted churn risk. The score is recalculated on a regular basis, so the at-risk population evolves over time as subscriber behaviors change. {% hint style="danger" %} This score is designed for businesses **with a subscription model** (streaming, SaaS, press, loyalty programs, etc.). It is not a general inactivity score; it is specifically calibrated to predict contractual cancellation intent. For other predictive churn scores, you can have a look here. {% endhint %} ## Why Use It? Retention is key for subscription model, because acquiring a new subscriber costs far more than retaining an existing one. By the time a customer has already cancelled, it is too late to act. With the Subscription Churn score, you can: * **Detect early cancellation signals** before subscribers take action, giving you a window to intervene. * **Trigger proactive retention campaigns** (renewal incentives, value reminders, exclusive offers) precisely when they are most needed. * **Avoid blanket retention campaigns** that erode margin by offering discounts to subscribers who were not at risk in the first place. ## How to Use This Score in Batch ### 1. Orchestration: Automate Retention When Risk Crosses a Threshold (Primary Use) The recommended setup. Because the score is recalculated regularly, subscribers move in and out of the at-risk population continuously. A trigger-based automation ensures they receive a retention message as soon as they cross your risk threshold. **How to set it up:** 1. Go to **Automations** in your Batch dashboard. 2. Create a new automation with an **Attribute Change** trigger: * Trigger condition: `Predictive score [Subscription Churn] crosses above [threshold]` (the right threshold depends on your business; your CSM can help you calibrate it) 3. Define your retention message: a personalized email or push highlighting subscription value, a renewal incentive, or an exclusive offer. 4. Add a **capping rule** to avoid re-entering the same subscriber too frequently (for example, no more than once every 30 days). 5. Activate the automation. From that point on, every subscriber who reaches the risk threshold automatically receives your retention message. {% hint style="info" %} [Use a **Control Group** on your automation](/guides-and-best-practices/orchestration/how-to-use-the-random-split-feature.md): exclude a random portion of at-risk subscribers from the retention flow and compare their cancellation rate over 30-90 days against those who received the message. This gives you a clean measure of the retention uplift. {% endhint %} ### 2. Segmentation: Periodic Retention Campaigns If you prefer to run manual campaigns rather than a continuous automation, you can use the score as a targeting filter on a regular schedule. **How to set it up:** 1. Go to **Campaigns** in your Batch dashboard and create a new campaign. 2. In the targeting step, apply a condition based on your configured threshold, for example: `Predictive score [Subscription Churn] ≥ 0.7`. 3. Optionally combine with a subscription end date filter to prioritize subscribers whose renewal is coming up soon. {% hint style="info" %} If you run this campaign regularly, save the audience as a named segment (for example, `[AI] Churn risk - subscribers`) so you can reuse and monitor it over time. {% endhint %} {% hint style="warning" %} In campaign mode, subscribers who become at risk between two sends will not be reached until the next scheduled campaign. The trigger mode is more effective for time-sensitive churn prevention. {% endhint %} ## Real-World Use Cases | **Industry** | **Example use case** | | ----------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Streaming & Media** | Subscribers whose engagement is dropping: automatically send a curated "you might have missed" digest or a personalized content highlight before their renewal date | | **Press & Publishing** | Subscribers at high churn risk: trigger a loyalty renewal offer (discounted rate, additional month) before they initiate cancellation | | **SaaS** | Users who have stopped logging in: send a personalized re-engagement sequence highlighting features they have not used, triggered as soon as churn risk crosses the threshold | | **Retail (loyalty programs)** | Loyalty members showing disengagement signals: trigger a "your benefits are waiting" reminder or a bonus points offer to reactivate their subscription before renewal | *** ## Related Documentation * [How to Measure the Performance of Batch AI Predict Scores](/guides-and-best-practices/batch-ai/how-to-measure-the-performance-of-batch-ai-predict-scores.md) * [Segments](https://doc.batch.com/getting-started/features/customer-engagement-platform/profiles/segments) * [Orchestration Targeting](https://doc.batch.com/getting-started/features/customer-engagement-platform/orchestration/targeting) * [Orchestration Analytics](https://doc.batch.com/getting-started/features/customer-engagement-platform/analytics/orchestration-analytics) * [Conversion Goal dedicated insights](https://doc.batch.com/getting-started/features/customer-engagement-platform/analytics/orchestration-analytics#conversion-goal-dedicated-insights) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://doc.batch.com/guides-and-best-practices/batch-ai/how-to-identify-and-retain-customers-at-risk-of-cancelling-their-subscription.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.