Discount launch pricing for the Data Product Leadership Community ends Sep. 30!
Participate in our live monthly Zoom webinars (recorded for on-demand access), connect over a 24/7 Slack, and get introduced to a new peer 1x1 every 2 weeks. Join my free Insights mailing list to get subscriber-only pricing before Oct. 1.
Get the coupon then apply for membership

Why software teams need to look beyond “user-centered” when referring to ML or AI-driven data products

As a designer, I used to say "user-centered"—a lot. It's terminology we now hear from non-designers now, people like many of you.

That's a good thing. But, I want you and your teams to think bigger.

For me, that "user-centered" descriptor is missing something about design, and particularly so if you are working on ML/AI systems.

Design can operate on multiple levels. I see at least three I want you to be aware of:

  1. user-centered - where we focus on the customers and end users of our product, what they want to do, how they want to do it, etc. We begin to drive more adoption, the first gate to increasing value.
  2. user-centered and business aware - everything above, but we're also using design to ensure value is created and business goals are met. Some of you may not see the distinction, particularly if you're an enterprise data team working with internal customers. The users/customers/business seems like one entity. Not exactly; and definitely not the case if you're at a software/product company. However, the main point here is that design needs to be operating on 2 levels now: creating wonderful UXs, while also creating business value—and balancing these two facets when they are sometimes at odds.
  3. human-centered - where we look at the users/customers, the business needs (they're also people!), and all the other relevant humans in the loop who may not be part of the design process but are affected by it. This is especially true as companies develop AI/ML systems trained on data sets for which the humans involved have no say; no recourse. They aren't users. Some of them are actually used. But, they are humans. Are you asking non-technical questions about your training data? Who isn't in the drafting room that should be? How can one really abuse this system since it doesn't have the intentional guide-rails of business logic? What recourse is there if a system is determining outcomes on wrong data?

I want you to be at #3, but if you're at #1, congrats—you're still making progress. Many of you data science and analytics pros are just getting started thinking about design, the last mile, and the UX of your solutions—and there is big opportunity as you're making the change from "data first" to "people first."

Omar Khawaja—my guest and head of BI from Roche Diagnostics—reminds us of this in the past episode of the podcast, which resonated with so many.

Now, if you're a design professional on this list, probably at a tech company or services firm saying, "we don't get no respect," "engineering runs everything," and "no time for research," you may start thinking more about #2; the business probably isn't seeing the value yet—even if users do.

And if you're working in ML/AI, I really hope you're soon to be focused on #3. Many of you on the data science side may think you're their to build models, but you're helping shape the culture too—intentionally or otherwise.

 

Photo by Jack Hunter on Unsplash

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

More Free Insights: