When it comes to working with communities, numbers are only the beginning. Finding insights, motivations, the factors that go into what participants and members are interested in, what they want - that requires going a few steps beyond simple data.

To be clear, we are talking about information. While data is something you can view and report on, information is data taken to the next level, a level where action can be taken and decisions can be made. For example, knowing there are eight people at a meetup is data. Knowing there are five developers, one organizer, a speaker, and a pizza delivery person means you know where to focus your energy (hint: it’s the pizza delivery person).

‍The question comes up then, how do we distill data to become information?

‍One way that is increasingly being used is applying the concepts of Machine Learning. This is the part where many people will take a step back and say, “Machine Learning? For communities of actual people? That’s outside the expected norm!” And that’s partially true - it’s not what most people are using to analyze their communities currently.

Many of the products and applications currently available for monitoring a community, whether it be a community of Open Source developers on Github or users for a particular application, just show counts. It may show some date ranges of activity or “last logged in” type of data point. With Machine Learning we can actually convert these data points into something useful.

‍It’s been said before, just providing a count is not very insightful.

‍With applications like Peritus.ai we can start to see insights. What are the parts of our application that our community appreciates? What are our Open Source contributors focused on most and why? How can we bring more people into our community and build it up?

A great example of this is when looking at parts of a group that are shining - the ambassadors. Just knowing you have a community of 5000 developers doesn’t let you know who the folks representing your brand or the community are. Machine Learning insights based on various activities linked to individual community members can help you discover your next Superstar.

Other benefits include automated information - no manual tracking or interaction. Often, community work is done by instinct. We find community members that are relatable or reflect something we are looking for - all based on our own perceptions. Those perceptions come with a certain level of bias. Machine Learning can remove that bias and focus on only the criteria important to building our best community.

Peritus is a step in the direction of making a more cared for, more equitable, community experience for Developer Relations practitioners, developer marketers, community managers, and anyone working with organized communities of professionals.