One trend in analytics and data science I am hearing from the top thought leaders in this space is that (finally!) companies are realizing that the biggest opportunities for data that lay before us is in the pursuit of new products or enhancing existing ones, as apposed to solely focusing on incremental operational improvements.

Gartner's CDO v4 definition is also all about product. 

Now, the products themselves may or may not be "data products," but this almost doesn't matter.

If you're a leader in analytics or data science, and you are entirely focused on "working the data" as the main focus of your daily efforts, you're never going to consistently create innovative solutions that wow your customer and make them rely on you as an expert in putting data to work for them in useful ways.  

So, what does innovation really look like?

Well, I know what it doesn't look like.

Here's what innovating doesn't look like when we're building solutions and products for humans:

• You think that 97 page "report" is helping a leader make a decision because it's chocked full of details and numbers
• Your model is 92% accurate but you have no idea who uses it to make decisions, when they use it, or how they use it, and whether it produced any meaningful downstream outcomes
• Your slide deck has lots of tables in it and "evidence" that should be obvious
• You think it's somebody else's job to make sure that data science and analytics initiatives produce value and get used
• You blame your customer or sponsor when they say "what do I do with this? This isn't what I asked for/needed. What does this mean?"
• You think your or your team's job is primarily to do math, statistics, or report building, mostly in isolation
• The first thing they saw from your team was likely digital (some software, an app, or report)
• Writing code or using software tools was the first step in the project/product
• There was little or no input from the users or stakeholder once they gave you their "ask" or "requirements"
• You gave 'em a model or chatbot when they asked for it, no questions asked. (Of course we can make one!)
• You definitely didn't question their request for a data-driven solution, or how they planned to use it
• Empathy and phrases like "so that?" aren't regular parts of your vocabulary or strategy and you definitely didn't ask "why."

So, why should you care about any of this?

Sure, some of you can bat the industry average of about 15-30% success rate on your AI, analytics, or big data project, and perhaps get away with that for awhile.

I'd argue that a 80% fail rate doesn't sound like a fun job, and eventually, leadership is going to expect results from data leaders. Your days may be numbered. Even moreso if you've got that VP, Director, or CxO title next to your name.

At some point, boards and CEOs are going to start expecting more when they toss those data-dollars in your direction.

Are you thinking, "it's not my problem?"

Well, here's what one leader in the data advisory space told me just the other day about data scientists and analytics leaders who think it's not their job to help make better products and decision support solutions:

"To be blunt - we've seen those folks who feel that way become unemployed."

However, it doesn't have to be that way.

If you want to become an indispensable ally to your product or business counterparts, learn about my consulting services, or take my seminar, Designing Human-Centered Data Science Solutions. You can learn the skills that will compliment your knowledge of data science, math, statistics, and analytics so you're consistently creating useful, usable, valuable decision support solutions. If you're not ready for that, you can join my free Insights mailing list or download my free self-assessment guide.

You're probably a problem solver at heart, but part of being a great problem solver, is being a great problem finder. This is where design comes in.