The Data Stack Show - The PRQL: How Important Is the Human Factor When Working With Data?

Episode Date: February 18, 2022

Eric and Kostas preview their upcoming show with Sean Halliburton of Warnermedia.

Welcome to the Data Sack Show prequel. We're going to talk with Sean Halliburton, who works for CNN. He was at Nordstrom before. Costas, one of the topics that we just tackled on the show we just recorded was clickstream data. And, you know, I don't have an encyclopedic knowledge of all the shows, but I don't think we've talked about clickstream data very much on the show. We've talked, we've had a very few conversations about real-time data and streaming. Why do you think that is? That really stuck out

to me that we haven't had a ton of conversations around sort of the clickstream real-time data use cases. I think the main reason why this is happening is the clickstream real-time data use cases. I think the main reason that this is happening is because clickstream data is just a use case of streaming data. And as we focus more, I mean, we talk more with engineers and data engineers, and the data engineers, I don't think that it matters that much if it is clickstream or some other kind of streaming data. At the end, the way that you have to handle the data is the same and you have like similar challenges. Now, is it different in the way that you capture the data with clickstream data? Yeah, it is.

But that's not that much, let's say, concern of the data engineer that much. I think it's more of a concern like of the data engineer that much. I think it's more of a concern of the product engineer who needs, I don't know, like SDKs or whatever they have to do to capture the clickstream data and submit this data. So I think that's the main reason that we haven't focused that much on it. I think the people that we talk with and ourselves, we just consider clickstream data as yet another case of data that we have to work with. Sure. Yeah, it's just interesting because I think we have talked a lot

about maybe the ETL side of things, which, you know, you can ETL clickstream data. So anyways, getting down a rabbit hole, what stuck out to you in the episode? I think it's something that our listeners who have listened to many of our episodes comes up a lot. And that's like the human factor at the end when it comes to working with data. If you remember, during our conversation, there was a word that was used a lot, and that was balance. You need to find the balance between, for example, how much you want to debug or how much tolerance for error you have,

or find the right way to communicate with stakeholders, both in your team and your internal customers. I think once again, we see that for data engineering and anything that has to do with data outside of the core and the hard technology itself that enables things, there is the human factor and the soft skills that are required to make these technologies work at the end and not cause problems. So I think that's what I keep. And it just reinforces something that we have talked a lot in the past. And it's very interesting to hear that also from him

because he has a very extensive experience in many different roles, both as an individual contributor, but also as a manager. I would agree. That episode was, to use a word that both you and Sean used in the episode, balanced. He had such good insights on the technical side, but always balanced that with considering the human side, both of this team and sort of the downstream customers of the data. So definitely check out this episode. We'll publish it after this one. Subscribe if you haven't, and we'll catch you on the next show.