While prediction results from AI Commerce Search are most commonly used
in retail websites, you can also use recommendations in email
messages. Recommendations in email are usually included as part of marketing
campaign email blasts, after-order emails, or abandoned cart emails.

## Choose a model type

The Recommended for You model can be a useful model for emails because it is designed to
deliver recommendations in use cases where a user hasn't indicated interest in a
specific item. It works best with logged-in users that have some browsing or
shopping history.

The Others You May Like model is typically used on product details pages rather
than for email campaigns. However, you can use it for emails to recommend
similar or complementary items to what a customer has recently viewed.

The Frequently Bought Together model is usually used on the add-to-cart page or the shopping cart
page, but can also be used for email.

The Buy it Again model is designed for use on most types of pages and can also be
used for email.

## Request requirements

In order to get relevant recommendations back, the [`userId`](https://docs.cloud.google.com/gemini-enterprise-cx/ai-commerce-search/reference/rest/v2/projects.locations.catalogs.userEvents.md.txt#UserInfo) field
must match the user ID that is associated with a logged-in user on your website
or app.

Ideally, the `visitorId` field should match the user's last session ID from the
site. Because the last session ID isn't always recorded, a random unique
identifier can also be sent as the `visitorId` field.

Frequently Bought Together and Others You May Like models also require product IDs in the
[`productDetails`](https://docs.cloud.google.com/gemini-enterprise-cx/ai-commerce-search/reference/rest/v2/projects.locations.catalogs.userEvents.md.txt#ProductDetail) section of the predict call. For Frequently Bought Together
models, the IDs can be items from a user's most recent purchase.

## Include static predictions in email

You can send static predictions by inserting data into an email template.

The `predict` method can be called in real time as emails go out, or the
recommendations can be retrieved individually and saved either as part of an
email template or in a database to be referenced later. To have the most up to
date recommendations, call the `predict` method as close in time as possible to
when the email is sent.

## Call the `predict` method for email

The AI Commerce Search [`predict` method](https://docs.cloud.google.com/gemini-enterprise-cx/ai-commerce-search/reference/rest/v2/projects.locations.catalogs.servingConfigs/predict.md.txt) takes some fields as
input and returns a list of product recommendations. `visitorId`, `userId`, and
`id` are used as input for most recommendations models.

Each prediction request for a user must be made individually, using code or
shell scripts.

The following is pseudocode for a predict request:

    for user in userlist:
      emailTemplate.recommendations = predict(user.visitorid,user.userid,user.lastorder.id,...)
    ...
    format email template as required

## Include dynamic predictions in email

Although static predictions can be quickly implemented, they can become outdated
more quickly than dynamic predictions. A user might not open the email for some
time, and better recommendations could be available by that time. More
sophisticated bulk email systems can add dynamic content by including a
reference to an image in an HTML email.

The following is example HTML for an image that includes a serving config
reference:

    <img src="https://example.com/recs/email?userid=47832&campaign=2020FALL&servingconfig=fbt>

Getting the recommendations back can be done with a Cloud Run function, or
any server-side app that can make the prediction request. The predict results
then need to be turned into an image. This can be done with a library, such
as [ImageMagick](https://docs.cloud.google.com/gemini-enterprise-cx/ai-commerce-search/tutorials/imagemagick).

Predictions are only made for opened, viewed emails. You can also cache the
results rather than creating a new image if the user opens the email again.
This reduces the cost of using recommendations for emails.

Because the link destination is typically hard-coded into the email, you don't
necessarily know which products will be shown. Redirecting clicks to individual
product pages requires additional configuration. You'll usually use a single
image for all recommendations, so using an image map can help you determine
which product was clicked.

## Quotas

If you plan to run a large batch of predict requests over a short period of
time, you should check your quotas. On your project's
[Quotas](https://console.cloud.google.com/iam-admin/quotas) page, check **Recommendation prediction
requests per project per minute** . You can request an increase by clicking
**Edit Quotas** on the quota's details page.

We recommend implementing
[exponential backoff](https://en.wikipedia.org/wiki/Exponential_backoff) to retry
requests after a delay. The service returns `HTTP 429` responses if quota is
exceeded.

## Evaluate results

URL tagging lets you track and evaluate the effectiveness of campaigns.
You can use [Google Analytics](https://developers.google.com/analytics/devguides/collection/protocol/v1/email) or another analytics
platform to add tracking parameters to links in emails and include that data in
reports. Tag recommendation clicks from emails to filter them so that you can
evaluate metrics.

If you A/B test multiple recommendations, we recommend doing so within a single
campaign. All standard [A/B testing guidelines](https://docs.cloud.google.com/gemini-enterprise-cx/ai-commerce-search/a-b-testing.md.txt) for websites still
apply.

## What's next

- [Create a serving config](https://docs.cloud.google.com/gemini-enterprise-cx/ai-commerce-search/create-configs.md.txt) for your model.
- Learn how to [pause and resume training](https://docs.cloud.google.com/gemini-enterprise-cx/ai-commerce-search/manage-models.md.txt#pause) for your model.
- When the model finishes training, [start requesting recommendations](https://docs.cloud.google.com/gemini-enterprise-cx/ai-commerce-search/predict.md.txt).
- Use [A/B experiments](https://docs.cloud.google.com/gemini-enterprise-cx/ai-commerce-search/a-b-testing.md.txt) to understand how AI Commerce Search impacts your business.