Too often in AI, we get caught up in what’s possible instead of focusing on what’s useful. Sure, we can "do AI" and learn a lot, but if it’s not solving a real problem, people are left asking why we poured so much time and money into it. And without real momentum, change never happens.
Reducing churn, increasing conversion, preventing mistakes, increasing engagement, reducing cost, speeding time to market...that is going to generate momentum and support for the change you are trying to make.
Why “Data-First” Isn't Always the Answer
When I reviewed AI frameworks from the big consulting groups like IBM, they emphasized the importance of collecting, organizing, and analyzing data to extract value. That sounds great—if you have ample time and budget. But after all that effort, you might realize there’s not much useful to do with the data. I prefer to start with the use case, then focus on the data with the most potential to drive results.
This isn’t to say that data isn’t critical—it absolutely is. But having the best data paired with a weak use case is like owning a Ferrari without gas. You’ve got something powerful, but it’s not going anywhere. Does it make sense to invest time and resources in cleaning, curating, and moving data that doesn’t support a meaningful application? Probably not. Sure, store it for future use—but don’t make it your top priority.
If you focus solely on easy-to-execute use cases, you’ll gain valuable experience in delivering AI projects. But here’s the problem: no one will care. You’ll have spent all that time and money, and people will be left asking, “What was it all for?”
What to Do Instead
Start with the problem:
How can we serve our customers better?
What will make their day easier?
How can we help their organization run more effectively?
Focusing on a valuable problem not only gives you direction but also helps you shape your data strategy. You’ll know exactly what data to collect, acquire, and curate because you’re solving for a clear, customer-focused use case. Having this long-term vision motivates your team, knowing that their efforts in gathering and managing data are aligned with real needs for the future.
Future Use Cases Drive Data Strategy
Product leaders should be focused on identifying use cases for the future. That way, the teams working on acquiring and managing data are moving in the right direction. AI is a powerful machine, but it’s not agile—it doesn’t turn quickly. You need to plan ahead. I’ve seen teams spend years collecting data, only to realize the schema (how data is organized) isn’t aligned with how they actually need to use it. That’s a painful lesson.
Why I Added “Usually”
I don’t love absolutes, so I’m giving myself some wiggle room. Sometimes, the data does need to come first—but most of the time, I’d rather start with the use case.
Image: A little watercolor I did on the quote from Abraham Maslow "it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail."
