3 types of Machine Learning Models and how to apply them to SaaS GTM

So you want to use machine learning to improve your GTM motion. But how does it even work and how can you apply it to your GTM challenges?

At a high level, ML is all about accepting data and outputting a prediction. This could be whether someone will convert or not, whether someone is like another someone or not, or the next best action to go from point A or point B.

If you’re thinking about applying ML to GTM, it’s important to understand the basics so that you can start thinking about how you might apply ML. Another important note is that, typically, ML is pretty complicated to get right, so although it’s great to understand the concepts, you’ll want the help of your data team.

There are 3 main categories of ML models that people like to talk about, and they are good at different things.

Supervised ML: Use this if you’re trying to predict an outcome using data that you already have on past outcomes.

Unsupervised ML: Use this if you’re trying to learn something new about the data set that you didn’t know before.

Reinforcement Learning: used more to teach a machine how to choose a most optimal path. Not *super* relevant for GTM motions

So let’s talk concrete examples.

Supervised ML: really interesting to use to predict conversions (binary => yes or no) or expected MRR (continuous)

Unsupervised ML: useful to find interesting segments of customers that look like each other

We’ll dive deeper into some of these topics in future blog posts!