Integration

Zembula + Algolia

Algolia builds recommendation lists for your products, and Zembula reads those lists for your email. Your own Algolia models pick the products that each email shows.

It's our business to grow yours

Book a demo

Zembula has built this integration.

What Zembula can read from Algolia

Algolia runs search and product recommendations for sites and apps. Your team trains Algolia recommendation models on your own data. Zembula reads the lists that those models make, so your email content can show the products that your own engine picked. Zembula reads four list types: Related products, Bought together, Looks similar and Trending.

Related products, Bought together and Looks similar give one list for each product in your catalog. Trending gives one ranked list for the whole catalog. A sync reads the lists each day and stores them in Zembula. When a person opens an email, Zembula shows the stored list, so the open adds no request to Algolia.

Your team connects Algolia in Zembula with three values: the Application ID, the host of the application and a Search API key. A key with the search permission is enough. Then you choose the index and the model for each list. Zembula can also read one indexed Record by its objectID, such as a product with the fields that your team stores in the index.

DataDetailLooked up byHow fresh
Related productsThe products that your Algolia model relates to one product.productdaily sync
Bought togetherThe products that people often buy with one product.productdaily sync
Looks similarThe products that look like one product.productdaily sync
TrendingOne ranked list of trending products for the whole catalog.whole catalogdaily sync
RecordsOne indexed item, such as a product, with the fields your team stores.objectIDset at setup

How the data arrives

  • API: A daily sync asks Algolia for the list of each product and stores the lists in Zembula.
  • ESP merge tags: The objectID reaches Zembula in the image URL, from a merge tag of your email service provider.
  • Product ID in the email HTML: Place the objectID in the email HTML, and Zembula can read the matching Record.

What you need

  • The Application ID of your Algolia application.
  • The host of your Algolia application.
  • A Search API key. A key with the search permission is enough.
  • A trained Algolia recommendation model for each list type that you want to show.

Limits to plan for

LimitWhat it means
The lists change once a day.A change in Algolia shows in your email after the next sync.
Algolia counts each request of the sync against your Algolia plan.Zembula shows the request count of each sync, so you can plan for it.
A list type needs a trained model in Algolia.A list type with no trained model gives no list.

Ideas for email

  • Show the products that Algolia relates to the product a person viewed.
  • Add a Bought together block to an order confirmation email.
  • Show the Trending list in a newsletter.

Last checked: 19 November 2026.

Performance marketing dials, inside email. For the first time.

See how we do it by booking a demo.