> ## Documentation Index
> Fetch the complete documentation index at: https://help.maestra.io/llms.txt
> Use this file to discover all available pages before exploring further.

# How to Create a Custom Recommendation Preset

Custom Recommendation Presets let you define complementary products through category pairs instead of leaning on statistical models. Once you say Category A should recommend items from Category B, every product in Category A gets paired with the item from Category B most often purchased alongside it. You can also layer on similarity parameters — price, manufacturer, or any other product attribute.

Maestra Platform offers two variations, both created the same way.

<CardGroup cols={2}>
  <Card title="Custom Recommendation Preset" icon="sliders">
    Category-based pairings for a specific product.
  </Card>

  <Card title="Custom Post-purchase Recommendation Preset" icon="receipt">
    Category-based pairings applied to each item in a customer's most recent order.
  </Card>
</CardGroup>

## Custom Recommendation Preset

Lets you set complementary products based on category relationships, for a specific item. Once you specify that Category A should recommend items from Category B, each product in Category A gets paired with the item from Category B most often purchased alongside it. Higher-priced items get more recommendations; lower-priced items get fewer.

* Best used on product detail pages.
* Works for identified and anonymous customers.
* Available through the API (recommendation widget) and email.

**Good to know:**

* **Refresh:** once a day.
* **Limit:** up to 5 instances per project.
* **Segment:** can be limited to a product segment.
* **Auto-filters:** checks the product's area; product external systems match by default.

## Custom Post-purchase Recommendation Preset

<Note>
  This preset only generates recommendations for customers with activity in the last 90 days. Everyone else will see recommendations calculated earlier (if the preset is more than 90 days old) — or no recommendations at all.
</Note>

The Custom Recommendation Preset applied to each product in the customer's most recently updated order, with recommendation volume proportional to price. Recalculates in real time based on the customer's orders.

* Best used for "Thanks for your order" and "Next purchase suggestion" flows.
* Identified customers only.
* Available through the API (recommendation widget) and email.

**Good to know:**

* **Refresh:** real time.
* **Limit:** up to 1 instance per project.
* **Auto-filters:** checks the product's area and brand (multi-brand projects); excludes items the customer has already bought.

## How to set it up

<Steps>
  <Step title="Open Product recommendations">
    Go to **Content → Product recommendations** and click **Add mechanic**.
  </Step>

  <Step title="Choose the preset">
    Pick **Custom Recommendation Preset** or **Custom Post-purchase Recommendation Preset**.
  </Step>

  <Step title="Name the preset">
    Enter a name and click **Continue**.
  </Step>

  <Step title="Configure general settings">
    * **Recommend for products** — the scheduled segment recommendations are built for (optional).
    * **Recommend from** — the scheduled segment recommendations are drawn from (optional).
    * **Recommend only from the same external system** — enabled by default; can be turned off.

    You can also configure **Manufacturer exclusions** in the corresponding block.
  </Step>

  <Step title="Set up category mapping">
    Select the category to map, then configure pairs and the number of recommended products per pair.

    When you specify that products from Category B should be recommended for Category A, each product in Category A is matched with the product from Category B that's most often purchased together with it.
  </Step>

  <Step title="Specify mapping settings">
    Mapping settings define how recommendations are sorted in the output. If **Price** is listed first and **Manufacturer** second, products matching both rank higher — and then are sorted by popularity in orders and views.

    Select the checkbox in the **Exact match** block to require an exact match on a field. For example, tick "Color" and only blue products will be recommended for blue products.

    <Note>
      The more fields you select, the fewer recommendations you'll get. Too many constraints can leave you with none at all.
    </Note>

    * **Primary category** — the category closest to the product.
  </Step>

  <Step title="Launch the preset">
    Activate it. The Product recommendations page shows the preset's status and the timestamp of its last update.
  </Step>
</Steps>