> 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-personalize-promotions-with-the-right-discount-level-for-each-customer.md). # How to Personalize Promotions with the Right Discount Level for Each Customer ## What Is the Promotion Recommendation Score? Batch AI Predict's **Promotion Recommendation** score predicts, for each customer, the most effective discount level to offer in order to maximize the probability of purchase while protecting your margin. Rather than a single float with the promotion sensitivity score, the output is a recommended promotional mechanic (for example, 10%, 20%, or free delivery) calibrated to each customer's individual sensitivity. {% hint style="warning" %} **Promotion Recommendation vs. Promotion Sensitivity: which one do you need?** * [**Promotion Sensitivity**](/guides-and-best-practices/batch-ai/how-to-personalize-promotions-based-on-each-customers-price-sensitivity.md) → You want to know *whether* a customer needs a discount at all. Focus: identifying who to include in a promotional campaign. * **Promotion Recommendation** → You want to know *which* promotional mechanic **suits** each customer. Focus: offer calibration and margin optimization. {% endhint %} ## Why Use It? Offering a 30% discount to a customer who would have converted at 10% is a direct margin loss. Offering 10% to a customer who needed 20% to convert means losing the sale. Promotion Recommendation solves both problems simultaneously. With the Promotion Recommendation score, you can: * **Avoid over-discounting** your most loyal and conversion-ready customers. * **Recover conversions** from price-sensitive customers by offering them the threshold they actually need. * **Run a single promotional campaign** while personalizing the offer depth for every customer in your database. ## How to Use This Score in Batch ### Personalization: Inject the Recommended Discount Level Dynamically (Primary Use) The core use case: use the recommended discount value as a personalization variable in your campaign template, so each customer automatically receives the right offer. **How to set it up:** 1. In your email or push template composer, add a discount variable block. 2. Use the Batch personalization syntax to inject the recommended discount: * `{{ profile.custom_attributes.batch_ai_promo_recommendation }}` 3. Build your template logic so the displayed offer (percentage, banner, CTA) adapts based on this value. 4. Every customer who receives the campaign sees a different discount level, calculated specifically for them. {% hint style="info" %} To measure performance, you can run an **A/B Content Personalization Test**: Variant A uses the Batch AI Predict recommended discount level, Variant B uses your current fixed discount. Compare conversion rate, average order value, and campaign margin. See the [Performance Measurement Guide](/guides-and-best-practices/batch-ai/how-to-measure-the-performance-of-batch-ai-predict-scores.md) for details. {% endhint %} ### Segmentation: Exclude Low-Discount Customers from Heavy Promotional Campaigns You can also use the recommended discount value as a segmentation filter to protect margin on your least price-sensitive profiles. **Example setup:** * Exclude customers whose recommended discount is below a defined threshold from a high-discount promotional campaign * Send them a non-promotional version of the campaign instead ## Real-World Use Cases | **Industry** | **Example use case** | | ------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | **Retail** | End-of-season sale: send each customer a personalized discount level (10%, 20%, or 30%) automatically injected into a single email template, based on their predicted conversion threshold | | **Beauty & Cosmetics** | Monthly promotional campaign: replace a fixed 20% discount for all with a personalized offer that protects margin on loyal customers while recovering conversions from price-sensitive ones | | **Travel** | Flash sale: personalize the discount depth for each booker profile, offering deeper reductions only to customers who demonstrably need them to convert | | **Media & Entertainment** | Subscription renewal campaign: calibrate the renewal discount individually to maximize the number of retained subscribers at the lowest possible promotional cost | ## 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) * [How to Personalize Promotions Based on Each Customer's Price Sensitivity](/guides-and-best-practices/batch-ai/how-to-personalize-promotions-based-on-each-customers-price-sensitivity.md) * [A/B Testing: Message Overview](https://doc.batch.com/getting-started/features/customer-engagement-platform/message/overview#a-b-test) * [Personalization in Batch](https://doc.batch.com/getting-started/features/customer-engagement-platform/message/personalization) * [Orchestration Analytics](https://doc.batch.com/getting-started/features/customer-engagement-platform/analytics/orchestration-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-personalize-promotions-with-the-right-discount-level-for-each-customer.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.