Webinar discussion: Delivering Real Value - What are data teams like, that are able to be pretty effective, impactful, and able to really make a difference?

Continuing the discussion from the Building Effective Data Science Teams Webinar:

Delivering Real Value

What are data teams like, that are able to be pretty effective, impactful, and able to really make a difference?

Our panelists for this webinar were:

  • Kobi Abayomi, Senior VP of Data Science at Warner Music Group
  • Gregory Berg, VP of Data Science at Caliber Home Loans
  • Elaine McVey, VP of Data Science at The Looma Project
  • Jacqueline Nolis, Head of Data Science at Saturn Cloud
  • Nasir Uddin, Director of Strategy & Inspirational Analytics at T-Mobile
  • Moderated by Julia Silge, Software Engineer at RStudio, PBC

Some discussion in the webinar:

Greg : I have a couple ideas. First, the ability to communicate effectively with the stakeholders. I mean Jacqueline, Elaine, Nasir, and Kobi all mentioned this - communicating with stakeholders. I don’t think that point can be hammered in enough - getting that relationship, communicating effectively, and keeping them in mind when developing things instead of developing things in isolation. It may seem like common sense, but it’s not always done in practice. Also, continually involve them in decisions throughout the process instead of going off into your own world, creating something and then coming back. The “here’s what I did for you” approach can come off negatively, instead of “let’s work on this together and solve this business problem.”

Another thing, from the organizational standpoint, is a mindset of being open to change. Just because this is how things have always been done doesn’t necessarily mean they always have to be done that way. Having that ability to change and being open to change is important.

The other characteristic - and I’ve seen this in several areas - is a diverse set of backgrounds for the team members. For instance, if everyone was like me and had a PhD in economics, focusing on econometrics, we’d communicate well and all be thinking of the same thing, but that doesn’t help the business. That doesn’t help my team. I’d want to include computer scientists, industrial engineers, etc. I want other disciplines that have a broader perspective, and that really opens up an open source of ideas. It’s not just the leader coming up with the ideas. If you have this diverse skill set, they’ll come up with great ideas on their own, working with the business, and it creates an environment of growth and value for the business.

Jacqueline : I agree with everything that was said, and I think a real truth that I’ve noticed when I’ve led data science teams, is that the team rarely fails because, “Oh, we had one data scientist and two data engineers. We should have had two data scientists and one engineer.” What usually fails is there’s not a clear focus. A team fails when they’re hired because “Hey, we have a lot of data, and, hey, you’re data scientists. You figure out what to do with that.” A team will fail if there’s not a clear focus.

To the point of flexibility, when the team recognizes that they aren’t finding value in this project, how quickly can they pivot to something better? This is partly people who are flexible and can do that and partly a business environment where the data science team can say, “It turns out that a churn model isn’t actually effective here.” And people say, “All right, that’s fine. Let’s move on to something else.” Instead of, “Well, why don’t you spend another year on it and then come back?”

Greg : To add on to that, I think it builds credibility if you can deliver negative results and explain that. Sometimes the business has an idea of what they want and they may be really excited, but it’s not how it works in reality. Having the ability to say, “No, this is not how it works” lifts the credibility of the department being the impartial observer and giving results, rather than just giving people what they want to see.

Julia : It’s interesting that you bring that up, because having to say no or explain a negative result, that the thing you wanted to do is no better than the way things were before, is something I have actually personally experienced as being quite challenging.