Stop Trying to Build a “Data Culture”

There’s a familiar pattern in the data world: we get caught up in the latest trend and lose sight of what really matters. One area that’s been given almost mythical status over the years is “data culture.” But in reality, many organizations haven’t taken the time to understand the culture they already have — let alone how to work with it.

I know many of you are working hard to introduce a stronger focus on data within your organizations, and I’m not here to dismiss those efforts. What I would like to do is help shift the approach slightly, so it lands better and delivers more impact.

Having worked in the industry for over 20 years, I’ve seen plenty of trends come and go. One that’s stuck around is the idea of building a “data culture.” But often, that ends up feeling like something separate, bolted on rather than built in. That’s the real issue.

Instead of chasing a standalone “data culture,” the focus should be on embedding data and AI into the culture you already have. Because unless it’s aligned with how your organization truly works, it won’t take hold, and we’ll keep seeing the same struggles repeated.

Many organizations avoid the hard truth and that is, culture isn’t a blank slate. It’s simply not something you rebuild every time a new initiative rolls in. It’s lived, it’s often messy, and it’s already shaping how your people think, decide, act, and perform.

I play the trumpet, and over the years I’ve played in both orchestras and jazz bands. So, here’s how I think about culture and data: trying to force in a brand new “data culture” is like dropping a jazz band into the middle of an orchestra with no rehearsal. There might be plenty of talent in the room, but without alignment, boy it’s just noise.

What’s the better move?

Infuse data and AI into the culture that runs your business today, and not the one you wish you had.

The Data Says It All: Culture Is Still the Biggest Obstacle

Don’t take my word for it, take the survey results that Randy Bean and Tim Davenport produce every year. According to the 2025 AI & Data Leadership Executive Benchmark Survey, companies are still struggling to embed data and AI in a way that actually changes how the business runs:

• Just 32.5 percent say they’ve established a Data & AI organizational culture in 2025.
• That’s down from 42.6 percent in 2024.
• In fact, over the past 5 years, most companies have barely moved the needle: 2021: 24.4 percent 2022: 19.3 percent 2023: 20.6 percent 2024: 42.6 percent 2025: 32.5 percent

Despite massive investment, talent acquisition, and strategy documents, most organizations are still stuck. And the No. 1 reason cited? Cultural challenges.

So no, the answer is not another committee, dashboard, or Centre of Excellence.

How Do You Actually Infuse Data & AI Into Your Business?

Here’s what it looks like when you treat culture as the engine, not the obstacle:

1. Start With Decisions, Not Dashboards

About a decade ago, I created something I called the “Decision Map.” Rather than starting with data, as everyone else was, I decided to flip the lens and start with decisions. I shared the concept widely, speaking about it at conferences, webinars, and industry events. At the time, it raised a few eyebrows. The data world was firmly focused on, well you know, the data. The Decision Map came about because one of my first big clients challenged me to think differently in this area, and it really helped me and them to look at it through a different lens.

Fast forward to today, and it’s great to see so many people now talking about “decisions” as the foundation for data and AI. That shift isn’t down to me, of course pioneers like Dr. Lorien Pratt and Lori Silverman have made major contributions in this space. I was just fortunate to have come across their thinking early on, and it deeply resonated.

Today, that early “Decision Map” has evolved into something more complete, which is my Data Strategy Canvas. But the principle remains the same, start with the decisions that matter, and build from there.

So, the question we need to ask: What are the real, painful decisions your teams have to make each week?

If data and AI don’t improve the confidence, speed, or quality of these decisions, then you are  simply playing with toys.

2. Understand the Frictions in the Culture You Already Have

In a lot of the corporate roles I’ve had over the years, I’ve seen the full spectrum when it comes to how teams relate to data. Some don’t trust the numbers at all, others worship them without question, and then there are those who go purely on instinct and won’t budge.

None of these are inherently wrong, to me they are just different starting points. But one thing we need to be careful about, especially in data teams, is the temptation to try and turn everyone into a data disciple. That’s not the job.

The real job is to make it harder for people to make poor decisions, and easier for them to make better ones. Data and AI should support that process, they are enablers, and not the North Star.

3. Build Around How Work Really Happens

Most C-suite leaders overestimate how cross-functional their teams really are. Strategy documents don’t make you cross-functional, the operating model and action does.

When I did my first proper data strategy back in 2002, though we didn’t call it that then, I led the data and insights team. One of the first things I did was ask the team to spend time with different departments across the business. Sit with them, watch how they work, understand how they make money, what their day-to-day processes look like, and how they actually use systems and data.

Back then, there was no AI in the picture. It was just about getting close enough to the business to understand how we could help. What that gave us was context. It meant we weren’t building reports in a vacuum (no dashboards in those days), we were solving problems in real workflows, with tools people already used, at the moments that mattered, albeit not with the tools people have now! We didn’t always get it right, but we had empathy for how people worked. That made all the difference.

4. Measure Outcomes, Not Inputs

I’ve been seeing a sudden surge on LinkedIn where people are now really seeing that dashboards aren’t the panacea that they were meant to be. Maybe, it’s due to the GenAI trend has now taken over. But here is the sad truth, no one cares how many dashboards you built, models you trained, or analysts you hired. I will amp that up and put it on repeat, as I know I often do!

No one cares how many dashboards you built, models you trained, or analysts you hired.

What they care about is whether there was an outcome or result linked to a business objective that is of vital importance, you know things like: Did churn drop? Did revenue go up? Did customer NPS improve? Did your teams move faster and feel more confident?

That is how you know it’s working and that’s how you know the culture is shifting.

5. Give Power to the Edge

If your frontline teams can’t act on the insight that you provide, then sadly and as you know because many people have spoken about this, you have just created a reporting factory.

Infusing data into culture means trusting people with it. Giving them the access, the training, the authority and yes, the space to get it wrong before they get it right.

Culture shifts when the person closest to the customer can make a better call, faster, because of the system you’ve built.

Evolve Your Culture

To create real impact, Data and AI need to be embedded deeply within the fabric of your business. They become powerful only when they shape decisions, influence actions, and give your people the tools to outperform. Forget the flashy posters, the catchy slogans, and the staged theatrics, they are just distractions that drain resources without delivering results.

If you genuinely want to transform your culture, focus on enabling your teams to make smarter, faster decisions. When you get that right, everything else will naturally align. That’s where true value lies.