> 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-target-the-right-customers-for-every-product-campaign.md). # How to Target the Right Customers for Every Product Campaign? ## What Is the Product Propensity Score? Batch AI Predict's **Product Propensity** score predicts, for each customer in your database, their likelihood to purchase within the upcoming days inside a specific and predefined product, category, or brand target. The output is a float between 0 and 1: the closer to 1, the stronger the predicted propensity. This score is **configurable**: you define the product target you want to focus on with your CSM, and Batch AI Predict builds a dedicated scoring model for it. Each configuration generates a distinct attribute on the customer profile. {% hint style="info" %} **Product Propensity vs. Product Recommendation: which one do you need?** * **Product Propensity** → You have a campaign on a specific *product or category* and want to find the best customers to target. Use this for segmentation: *"Who should I contact for this campaign?"* * **Product Recommendation** → You want to find the best products to show to every customer. Use this for personalization: *"Which products should I put in this customer's template?"* {% endhint %} ## Why Use It? Most product campaigns go out to the full database by default. The result: low relevance, high unsubscribe rates, and wasted budget on customers who have no interest in what you are promoting. With the Product Propensity score, you can: * **Reduce your campaign volume** by targeting only high-propensity profiles. * **Increase your click and conversion rates** by reaching more receptive audiences. * **Protect your CRM pressure** and reduce opt-out risk across the rest of your base. ## How to Use This Score in Batch ### 1. Segmentation: Target Only High-Propensity Profiles (Primary Use) The core use case: use the score as a targeting filter directly in your campaign. You start from a product or category and identify the best audience to reach. **How to set it up:** 1. Create your campaign in Batch. 2. In the targeting step, apply a condition based on your Product Propensity score: `Predictive score [Score name] ≥ X` {% hint style="warning" %} **Choosing your threshold:** Batch recommends a default threshold that balances revenue impact and engagement quality. You can adjust it based on your campaign objective: a lower threshold increases your reach for broader revenue impact; a higher threshold targets only the most receptive profiles for better engagement rates. Your CSM can help you calibrate this. {% endhint %} {% hint style="info" %} If you plan to reuse this audience across multiple campaigns or want to monitor how it evolves over time, you can save it as a named segment: for example, `[AI] High propensity - [Category name]`. {% endhint %} {% hint style="info" %} You can combine the propensity score with other targeting criteria (recency, loyalty tier, last purchase date) to further refine your audience. {% endhint %} **Recommended test**: Run an **A/B/C Targeting Comparison Test** to compare the Batch AI Predict targeting against your existing business rules (e.g., "customers who purchased in this category in the last 6 months"). See the Performance Measurement Guide for details. ### 2. Orchestration: Go Further with Trigger-Based Campaigns Once you are comfortable with one-shot segmentation, you can take it a step further: use the score as an entry condition in an automation, so the right customers are reached automatically. For example, as soon as a customer's propensity crosses your threshold, regardless of any campaign launch. **Example setup:** * Automation entry condition: **attribute change trigger** on `Predictive score [Score name]` crossing your threshold * Action: send a targeted message to customers the moment they become high-propensity; no manual segment refresh needed * Result: no high-potential profile is missed between two score recalculations {% hint style="info" %} This pattern works well for always-on automations (e.g. a standing campaign for a category or brand) rather than one-shot sends; the automation runs continuously and catches each new high-propensity customer as they appear. {% endhint %} ## Real-World Use Cases | **Industry** | **Example use case** | | ------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | **Retail** | Seasonal promo (coats for winter, etc.): instead of blasting the full database, target only customers with high propensity for the relevant category; reducing volume while improving conversion rates | | **Media & Entertainment** | New series release: reach only subscribers with propensity for the relevant genre (action, documentary, etc.) and protect the rest of your base from irrelevant pressure | | **Travel** | Always-on automation: use an attribute change trigger to automatically notify travelers the moment their propensity for in-destination activities crosses the threshold; no manual refresh needed | *** ## Related Documentation * [How to Measure the Performance of Batch AI Predict Scores](https://www.notion.so/33d56eb49a8480b6941ceb021c0c6716) * [Segments](https://doc.batch.com/getting-started/features/customer-engagement-platform/profiles/segments) * [Targeting](https://doc.batch.com/getting-started/features/customer-engagement-platform/orchestration/targeting) * [Omnichannel Trigger Automations](https://doc.batch.com/getting-started/features/customer-engagement-platform/orchestration/trigger-automations?q=target#analysing-automation-with-analytics) --- # 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-target-the-right-customers-for-every-product-campaign.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.