that looked great on the surface but didn’t really say anything?

When I first attempted to make sense of my dataset one Saturday afternoon, building a dashboard seemed like the next reasonable step in my data science journey.

I’d binged enough YouTube tutorials to think I knew what a “good” one should look like, probably something with a clean layout and maybe a few filters on the side.

With all that, I jumped right in.

I made a structure of how I wanted it to be and laid out components for my dashboard, but when I finally pieced it all together, something felt off.

I stepped back to look at it, literally. I walked across the room and studied it from different angles. We all do this, right?

After a few long looks, I couldn’t explain what story the dashboard was actually telling.

And don’t get me wrong, it was pretty decent for a first attempt. But looking at it felt like watching a group of people all talk over each other.

I had squeezed in different chart types—bar charts next to pie charts next to line graphs—all fighting for attention on one screen. Each chart had something interesting to say, just not in a way that added up to a clear point.

Later that evening, my stomach sank as I sat there staring at my screen, the blue glow reflecting off my coffee mug. If my own dashboard couldn’t connect with me, how could I expect it to connect with anyone else?

I started reading about why some dashboards fail to connect with people. I stumbled across a Harvard Business Review article that explained how many dashboards fail to drive real decisions because most analysts focus too much on looks rather than clarity.

It mentioned something about “chart junk”, just decorative elements that don’t add meaning.

That hit home. Ouch.

Look, data storytelling isn’t just about explaining insights. Rather, it’s about helping people see what you saw in your analysis and explaining it in a way that makes sense to them.

This article isn’t about the technical side of building dashboards; there are already countless tutorials that can teach you that.

Rather, it’s about the parts we often overlook: how dashboards communicate meaning and intent. I’ll also share the mistakes and lessons that changed the way I see data after building my first dashboard.

Why My Dashboard Looked Right but Felt Wrong

It took me a bit of humility to admit that the problem wasn’t the design.

It was me.

I was trying to tell a story I hadn’t actually discovered yet.

I began to see that data really isn’t the story itself; instead, it’s kind of like the language we use to tell one. And like any language, meaning comes from how we choose to arrange it.

That’s when I figured out I needed to pause before building anything and ask myself a few key questions first. I call them the three Ws:

• Why does this data matter?
• Who am I designing for?
• What question am I really trying to answer?

Those simple questions changed everything. My dashboards stopped being just visuals and started feeling more like actual conversations.

It took me a while to realize the problem wasn’t the tool or even the dataset.

It was the way I approached the story.

I had spent so much time trying to make the dashboard look right that I never stopped to ask what it was actually saying. It was like that moment in The Matrix when Neo finally sees the code. Once that thought crossed my mind, I knew I had to start over.

(If you want to dive deeper into dashboard design principles, this guide is solid.)

Building Again, but Differently

When I came back to the project, I decided to start over. But this time, I didn’t rush to open my visualization tool just yet. Which felt weird, honestly. My fingers were itching to click something.

I sat with the data for a bit, trying to understand what it was really saying and how I could guide that story toward interactive visuals.

Something about slowing down felt right. I started noticing stuff I didn’t see earlier, mainly small details that felt like they shouldn’t matter, but actually did.

Instead of trying to show everything, I decided to focus on one idea and build around it. For example, I had all these sales metrics sitting in front of me, but I picked one question that stood out:

Why were monthly sales dropping even though customer sign-ups were increasing?

That shifted everything. Suddenly, my visuals weren’t fighting for attention. Instead, they were working together to tell the same story.

As I went along, the less I added, the clearer everything became. I removed a few unnecessary charts and added brief notes to explain what certain numbers meant.

I added a simple annotation that said “Drop-off point” with an arrow pointing to where things started declining. No fancy design, just clarity. It wasn’t perfect, but it felt a lot more intentional.

I spent three days building the first version. The second one? Six hours.

Six.

Not because I rushed, but because I finally knew what mattered.

When I shared it, people didn’t just nod politely. They leaned closer, asked thoughtful questions, and also tried to guess what might be driving the trends. One person even pulled out their phone to take a picture of it.

It felt different, in a good way. Not gonna lie, I felt pretty proud.

Looking back, that moment changed how I approached projects afterward. I began to see dashboards less as something to display and more as a way to translate what I was seeing, and help others understand it.

Sometimes I still catch myself wondering if I’m doing it right, but maybe that’s the point. Maybe storytelling with data isn’t about getting it perfect.

Perhaps it’s about slowing down long enough to ask, what story am I really trying to tell here?

What I’d Tell My Past Self

If I could go back to that first attempt, here’s what I’d tell myself:

Start with pen and paper before opening the tool. Sketch out the story first. What’s the beginning, middle, and end? You don’t need software for that.

Delete one chart for every two you add. If it doesn’t directly support your main point, it’s just a distraction. Be ruthless with what you put in.

Read your dashboard out loud. If you can’t explain it in one breath, simplify. Your audience won’t have more patience than you do.

These simple rules have saved me countless hours and prevented me from creating more cluttered dashboards that look busy but say nothing.

I believe every dataset has a voice, but it takes patience to listen closely enough to hear what it’s really saying. And trust me, once you do, everything from the visuals to the insights starts to align with purpose.

Conclusion and Takeaways

When I first started, I wanted to prove that I could build something great. But by the end? Turns out the best dashboards aren’t the flashiest ones. They’re the ones that make people pause and say, “Oh. I get it now.”

That project taught me something I didn’t expect: data storytelling is less about the data and more about empathy.

There was this satisfying click when everything finally made sense—not just for me, but for everyone who looked at it. That feeling of connection, of being understood, made all the rebuilding worth it.

Now, whenever I open a new dataset, I remind myself of that lesson: start slow, listen closely, and build with intention. Sometimes I still mess it up. But at least now I know what I’m aiming for.

The goal isn’t to impress, it’s to connect.