I started Enterprise DNA by making Power BI videos. Just me, a screen recorder, and whatever I had learned that week about data visualization and DAX.

That was the beginning. Today, over 220,000 professionals have used the platform. We have a full course library, certifications, an AI-powered tool called Mentor, a community of data professionals, and a services arm called Omni.

I want to share the real story. Not the polished version. The version with the mistakes, the wrong turns, and the things I would do differently if I started over.

Starting with YouTube and hoping for the best

In the early days, my entire strategy was “make good content and put it on YouTube.” That was it. No business model. No customer research. No growth strategy. Just a belief that if the content was good enough, people would find it.

And here is the annoying thing. It kind of worked. The Power BI community was hungry for quality tutorials. There were not many people making them. The videos found an audience quickly.

But “kind of worked” is dangerous because it lets you avoid the harder questions. Questions like: who exactly is this for? What problem am I solving? How does this become a real business?

I rode the YouTube wave for a while, building an audience but not really building a business. The lesson I wish I had learned earlier: content is a great acquisition channel, but it is not a business model.

The transition from content creator to platform builder

At some point, I realized that YouTube views do not pay the bills in any meaningful way. Not at the scale I was operating at. I needed to build something people would pay for.

So I started building courses. Structured, in-depth courses that went far beyond what a YouTube video could cover. Power BI fundamentals. Advanced DAX. Data modeling. Analytics.

This is where things got real. Building a course is hard. Building a platform to deliver courses is harder. Building a business around all of it is harder still.

I made every mistake you can make. I built the platform before I fully understood what users wanted. I tried to do everything myself for too long. I underestimated how much customer support an education platform requires. I overestimated how quickly revenue would scale.

The first year of paid courses was humbling. The content was good, but the business operations were a mess. I was the instructor, the customer support team, the marketing department, and the developer all rolled into one.

What saved me was the community. Even when the platform was rough, people stuck around because they were getting real value from the content and from each other.

Why community matters more than content volume

This was one of the biggest lessons and one I almost missed.

In the early days, I was obsessed with content volume. More courses. More videos. More certifications. I thought the path to growth was having the biggest library.

It was not.

The businesses and individuals who stayed longest and got the most value were the ones who engaged with the community. They asked questions, shared their work, helped other members, and built relationships with fellow data professionals.

Content gets people in the door. Community keeps them there.

I watched members who completed all our courses but never engaged with the community drop off within six months. And I watched members who completed two courses but were active in the community stay for years.

The lesson: if I were starting over, I would build the community first and the content second. Not the other way around.

The mistake of trying to be everything to everyone

This almost killed us.

At one point, I was trying to serve complete beginners who had never opened Excel, advanced data engineers building production pipelines, business analysts who needed Power BI, data scientists who wanted Python, and everyone in between.

The course catalog was growing, but the experience was getting worse. Beginners were overwhelmed by advanced content in their feed. Advanced users were annoyed by basic material. Nobody felt like the platform was built specifically for them.

The turning point was when I sat down and looked at the data. Who were our most engaged, most satisfied, longest-retained users?

The answer was clear. Business professionals, usually in mid-career, who needed practical data skills for their jobs. Not aspiring data scientists. Not complete beginners. People who were already working in business and needed to get better with data and analytics.

When I focused on that audience, everything got easier. Course design got clearer. Marketing got more targeted. Support got simpler. We stopped trying to be the platform for everyone and became the best platform for that specific person.

Revenue went up. Churn went down. Satisfaction improved. The lesson was painful but obvious in hindsight: focus beats breadth every time.

Building the expert network

For the first few years, I taught everything myself. Every course. Every webinar. Every Q&A session.

That does not scale. And honestly, it does not serve the learner well either. I am good at certain topics, but nobody is the best at everything.

Building the expert network, bringing in other instructors and specialists, was one of the best decisions I made. It took the platform from “Sam’s courses” to a genuine learning community with diverse expertise.

But it was harder than I expected. Finding people who are both deeply knowledgeable AND good at teaching is rare. Technical expertise and the ability to explain complex concepts clearly do not always go together.

We went through a lot of trial and error. Some instructors were brilliant but could not simplify their explanations. Others were great teachers but did not have the depth that our audience needed.

The ones who worked best were practitioners. People who used these tools in real businesses every day and could teach from experience rather than theory. That became our hiring filter: do you use this in your actual work?

The technology decisions that mattered (and the ones that did not)

I spent way too much time and money on technology decisions that turned out to be irrelevant, and not enough on the ones that actually mattered.

What did not matter: Which video hosting platform we used. Which exact LMS features we had at launch. The specific design of the learning interface. Whether our certificates looked fancy. We agonized over these decisions and most of them could have been swapped out later without anyone noticing.

What mattered enormously: Search and discovery. Could people find the right content quickly? This made a massive difference in engagement. Recommendation quality. Did the platform suggest the right next step? Community infrastructure. Could people connect, ask questions, and share work easily?

The technology that matters is the technology that reduces friction between the learner and the learning. Everything else is decoration.

When we built Mentor, our AI tool, this lesson was front of mind. We did not build it to be technically impressive. We built it to be immediately useful. Can you ask it a question about your actual data problem and get a useful answer? That was the bar.

Why we expanded into services

This is the part of the story that surprises people the most. Why would a data education company start deploying AI agents, building apps, and providing AI advisory for businesses?

The answer is simple. Our users told us to.

Not in those exact words. But the pattern was unmistakable. Thousands of conversations over years, all pointing in the same direction.

“Sam, I learned all this. I understand it. I know what I need to do. But I do not have the capacity to actually do it.”

I heard it from property managers, logistics companies, dental practices, accounting firms, marketing agencies, consultants. Every industry, same problem.

They had the knowledge. They did not have the operational capacity to deploy it.

At first, I thought this was outside our scope. We are an education company. We teach skills. The deployment is on them.

But the more I heard it, the more I realized that if we did not bridge this gap, all that education was just creating more informed frustration. People who could see the potential clearly but could not reach it.

That is why we built Omni. Three services, Ops, Apps, and Voice, each addressing a different piece of the operational gap. We design, build, and manage AI solutions so that our users can actually deploy what they have learned.

From the builder’s perspective, it was a scary expansion. We were going from a pure digital product (courses) to a service business with real operational complexity. Different economics. Different hiring. Different customer expectations.

But it was the right move. And it only works because we built it on top of the education foundation. We understand what businesses need because we have trained 220,000 of them.

What 220,000 users teaches you about what people actually need vs what they say they want

I will end with this because it is the lesson that ties everything together.

What people say they want: more courses, more content, more certifications, more tools.

What they actually need: the ability to apply what they learn to their real work, support when they get stuck, and someone to help them deploy it when they do not have the capacity.

Early on, I listened to what people said they wanted and built more content. Engagement would spike and then plateau. I would build more content. Same pattern.

When I started focusing on what people actually needed, the application, the support, the deployment, everything changed. Retention improved because people were getting real results, not just completing modules. Satisfaction went up because the platform was solving their actual problem, not just teaching them about it.

People do not really want courses. They want to be better at their jobs. They want their businesses to run more efficiently. They want to stop missing opportunities because they do not have the right skills or the right systems.

Education is a means to that end. But it is not the only means. Sometimes what someone needs is not another course. It is an AI agent that handles their email while they focus on the work that matters.

That insight changed Enterprise DNA from a course platform to a company that helps businesses actually use data and AI. And I wish I had understood it from the beginning.

What comes next

I still make content. I still love teaching. EDNA Learn is not going anywhere.

But the vision is bigger now. We are building something that takes a business from “I do not understand data” all the way to “AI agents are running parts of my operations and my team manages them confidently.”

That is the full journey. And it took 220,000 people to teach me that the journey does not end at education.

If you are just starting that journey, EDNA Learn is the place to begin. If you are further along and ready to deploy, Omni is here. Either way, we have probably learned something from the 220,000 people who came before you that can help.

Come join us and find out.