In SaaS, machine learning has become an essential component to many different products. Whether it's automating responses to inbound sales queries, identifying expense reports for audit, or surfacing anomalies in data, machine learning improves workflow software. To date, most software imbued with machine learning reduces costs rather than increase revenues.
Why is this the case? Because machine learning is focused on efficiency gains.
First, to train a machine learning model requires large amounts of data. Repetitive workflow processes produce this data. With enough data, a machine learning model can be applied to automate the workflow. For example, it's straightforward to build machine learning models to automatically answer password reset questions.
Second, automation reduce costs because they eliminate the need to hire additional people to handle more work. A bot replies to login reset questions with an automated response in a link to a knowledge base, freeing customer support reps to focus on more challenging questions.
Third, revenue-generating activities are often unpredictable and novel. A new marketing campaign that positions a business intelligence product in a novel way can deliver a 30% increase in inbound lead volume. We haven't found a way, yet, to have computers do those things for us.
Perhaps that will change in the future. Generative machine learning - the software that composes music or writes code - could “imagine” a constellation of new marketing campaigns or sales pitches. Coupling machine generated ideas with the efficient automated testing systems we're developing today might unlock revenue growth opportunities.
By and large, the most frequent applications of machine learning in SaaS today are efficiency applications - automating the high-volume rote processes and reducing costs. Consequently, if you looking to build a machine learning based SaaS company, find a really expensive internal process and automate it.
Helping workers do more with less is always a compelling value proposition, provided the savings are large enough.