Welcome to What To Do With It — practical AI and data strategy for leaders who want results, not reports.
I’ve had the same conversation dozens of times.
A business leader — sometimes a CEO, sometimes a CDO, sometimes a VP who’s been handed the AI mandate and isn’t sure what to do with it — sits down and tells me about their data situation.
They have dashboards. They have a data warehouse. They’ve invested in Snowflake, or Power BI, or whatever the consultant recommended three years ago. They might even have a small data team.
Then they say some version of the same thing:
“We have all this data. We’re just not sure we’re getting everything out of it that we should be.”
That sentence, in one form or another, is why I started this newsletter.
Because that feeling — the sense that there’s more value sitting in your data than you’re currently capturing — is almost universal. And it’s almost never a data problem.
The bottleneck almost never lives where you think it does
When companies feel like they’re not getting enough from their data, the instinct is to look at the data itself. Maybe we need more of it. Maybe it’s not clean enough. Maybe we need a better tool.
So they invest in another platform. They hire more data engineers. They build more dashboards.
And six months later — same feeling.
Here’s what I’ve learned after a decade of building data teams and AI solutions across multiple industries: the bottleneck is almost never the data. It lives in the gap between the insight and the decision.
Data tells you what happened. It doesn’t tell you what to do. That translation is where most companies are stuck — and it’s the step that gets skipped almost every time.
That gap shows up in a few predictable patterns. See if any of these sound familiar:
– The data exists, but nobody trusts it. Two teams pull the same report and get different numbers. So everyone defaults to their own spreadsheet — and the dashboard collects dust.
– The insight lands too late. The weekly report shows up on Monday. The decision it was meant to inform was made on Friday. So the data becomes retroactive justification, not actual input.
– The decision-maker can’t interpret what they’re looking at. The chart is technically correct, but it doesn’t speak the language of the person who needs to act on it.
– There’s no clear owner of the “so what.” The data team delivers the numbers. Everyone waits for someone else to say what they mean. Nothing changes.
These are organizational and process problems. They don’t get solved by a better dashboard. They don’t get solved by a newer model. And they don’t get solved by AI — at least not automatically.
They get solved when data, AI, and decision-making are connected by design. When the system is built with the decision as the starting point, not the data.
That’s a different way of thinking about this. And it’s the lens I’ll bring to every issue of this newsletter.
What AI actually does — and doesn’t — fix
AI is genuinely one of the most powerful tools I’ve worked with in my career. At CoreTrust, we’ve deployed AI agents that have transformed workflows that used to take days into processes that run in minutes. I’ve seen what it can do when it’s implemented well.
But I’ve also seen the other side.
I’ve seen companies deploy AI chatbots and call it an AI strategy. I’ve seen six-figure pilots that never made it past the demo stage. I’ve seen teams spend months building models on data nobody trusted, for decisions nobody could define.
The pattern in every failure is the same: AI was treated as the solution to a problem that hadn’t been clearly named yet.
AI amplifies what you already have. If you have clear decisions and trusted data, AI makes your organization faster and sharper. If you have murky decisions and messy data, AI makes you more efficiently confused.
The good news is that clarity is achievable. The companies I’ve seen get the most from AI aren’t the ones with the biggest budgets or the most sophisticated tech stacks. They’re the ones that started with a clear answer to a deceptively simple question:
What decision are we trying to make faster, better, or more consistently — and what data does that decision actually require?
Everything else follows from that.
What you can expect from “what to do with it” newsletter
Every issue of What To Do With It will be built around one idea, framework, or real-world lesson from the front lines of AI and data leadership.
I’m not going to summarize the latest AI research papers. There are plenty of people doing that. What I’ll bring is the practitioner perspective — what this actually looks like when you’re trying to implement it inside a real organization with real constraints.
Here’s what’s coming in the next few issues:
– How to build a data culture that actually sticks — and why most companies start in the wrong place
– The 5-question framework I use before any AI investment to filter signal from hype
– What “data trust” really means, how to diagnose it, and the fastest path to fixing it
– Building a data team from scratch — what I got wrong the first time and what changed
– How to talk to your CFO about AI ROI (without getting laughed out of the room)
If any of those land for you — you’re in the right place.
One thing to try this week
Before your next data or AI conversation — whether it’s a vendor demo, a team meeting, or a budget review — write this question at the top of your notes:
“What specific decision will be different because of this?”
Not “what insight will we get.” Not “what data will we have access to.” What decision. Who makes it. How often. And what they’re doing today without this information.
If you can answer that question concretely before the conversation starts, you’ll run a very different meeting. And you’ll spot very quickly whether the thing being proposed is a real solution or a solution looking for a problem.
Try it once and see what it surfaces. I’d love to hear what you find
Glad you’re here. Let’s get into it.
— Derek Wilson
Sign up for the what to do with it newletter
Follow me on LinkedIn
