First of all, it’s so important to recognize that in B2B SaaS, you often don’t have enough data to rely on machine learning models in the same way that B2C companies do. That being said, there are some fun applications of ML that you can apply in B2B SaaS. In particular, I’m a fan of using ML for more exploratory analysis, throwing away the expectation that the model will be super accurate or predictive (that would be… black magic).
Use Case 1: Probability of Converting OR Churn Binary classification models are quite nice for predicting whether or not an entity will achieve a goal. Common use cases are if someone converted or churned.
Use Case 2: ARR Forecasting Regression models are quite simple to understand and a good jumping off board for forecasting.
Use Case 3: Customer Segmentation Clustering algorithms are useful in breaking customers up into Clusters. One weakness is that you do have to figure out how many Clusters you want, but one benefit is that you don’t need to know this beforehand!
Use Case 4: Recommendations This is a bit related to clustering, mentioned in Use Case 3, but you can actually make recommendations by looking at other customers who look alike and recommending what they bought to other similar customers. So basically, the assumption is that if someone is in the same “Cluster” they also might buy the same things.
So, don’t believe it if anyone tells you ML is a magic bullet, but there are some really nice applications of it in the GTM world!
