Forecast Sales to Trigger Marketing Actions on your Shopify Store

Optimize Customer Churn on your Shopify Store using Shopify Mail

Ever wanted to notify your consumers at the right moment, with the right product recommendation to optimize your company? 

Doing email marketing is complicated. Sesigning newsletters and recommending deals by email is hard and takes lots of efforts. 

With our solution, you can predict WHO will buy WHAT and WHEN to optimize your store and automatically send personnalized emails with the right product at the right time to the right person to keep them in the loop. 

Finally: predict WHO will buy WHAT and WHEN will each of your consumers buy. 

Our algorithm can predict WHAT and WHEN will each of your consumers buy, in a personalized fashion, based on their transaction history and your product catalogue.

In this sense, our algorithm can predict the following things for each consumer:

  • WHAT: The probability of the consumer consuming this product. Optional: also predict the quantity.
  • WHEN: The amount of days in which the consumer may buy again. Note: for specific applications of this algorithm, it would be possible to adjust the time frame.

Providing the description of each product in the transaction history and in the product catalogue will help the algorithm using advanced Natural Language Processing (NLP) techniques, in combination with our advanced time series forecasting solutions.


First, integrate with your data source. You must feed our algorithm with your customer data to be able to do the predictions. We don't resell your data to any third parties, and your data is considered confidential information. 

You ideally have at least 3x more transactions than products. You should also have at least 100 transactions in your system's history, although it'll be possible to generate predictions for new consumers having done only 1 transaction at least, as the transactions of similar consumers in the past will help in predicting the right thing. Finally, we estimate that you should ideally have less than 5000 products in your catalogue for the predictions to work well.

The algorithm can be retrained overnight to consider the new transactions of the day, and will be able to refresh the predictions.

The predictions

The predictions, once gathered, can be analyzed from a “per product” or from a “per consumer” standpoint, depending on which dimension you slice the generated results.

It is then possible to use triggers to launch events just before the prediction consumption time, for each of your consumers - for instance, emailing a newsletter at the right time with the right products listed in it (think IFTTT, Zapier, and

Integrations with your business consumer data

To generate predictions, we need to connect to your data sources.

We yet have built custom data connectors for our clients to interface with their business' data. 

Our next step is to integrate with Shopify connect with E-Commerce stores. 

Integrations with other Shopify tools for generating triggers

Once you generate predictions, they can trigger automated actions, such as sending emails, text messages, slack messages, and any other forms of automated reminders or messages.

Although we have yet built custom integrations for our first clients' needs, our next step is to integrate with Shopify's workflow triggers and Shopify mail:

Request a demo now!

Email to request a demo, or directly book a meeting