How to act on AI Predictions With Zapier ?

VincentSider

Hello,

I am new to DSS and this forum so apologies if the question has been asked. This is related to Zapier (I could only find one post about it in the forum)

I am wondering how to deploy a model to Zapier to monitor a customer CRM and predict churn and then take relevant actions, something along the lines of :

(1) connect Zapier to a dataset that has our customers i.e. hubspot or some other CRM, or even just Google sheet

(2) Webhook with a match up of the column names in the data with the column names used in the model

(3) We now have a Zap, which sends data from our customer CRM to our churn prediction model, and we’ll get the probabilities of churn as an output. With these predictions in Zapier, we can now do anything we want with them: Send them to a Slack channel, send an email to our team, add the prediction to our CRM, or anything else.

Any views ?

Vincent

Mark_Treveil

The standard approach would be to deploy your model as a REST API endpoint (see API Deployer) and call it from Zapier. Using Custom EndPoints you can support up any data lookups or other bespoke preprocessing.

Alternatively you could export the model and run it elsewhere. We support JAR file exports as well as PMML and for Deep learning models, ONNX.

VincentSider

Thank you @Mark_Treveil