From Dashboards to Decisions

May 12, 2025

Data Dashboards: True Windows Into a Business

Data dashboards are one of the clearest lenses into how a business thinks, moves, and reacts.

They’re used differently by every role in an organisation.

• Execs might glance and ask, “Are we okay?”
• Analysts may deep-dive for hours.
• Operational teams look for quick checks. And some people- frankly- see a wall of charts and panic.

Not everyone is fluent in the language of data. Some people instinctively scan for patterns. Others sit quietly, unsure what they’re looking at- and too afraid to ask. That’s not a capability issue. It’s just the reality of how diverse our decision-makers are.

Data dashboards aren’t about making sense to everyone instantly. They’re about making performance visible. But when they don’t provide that window, they fail.

And in today’s noisy world, people don’t just want context- they want clarity. The reason and the recommendation….all in one place…. there and then.

That’s where AI is starting to step in more and more.

Highlighting what matters. Summarising what changed. Bringing the most important signals to the front of the room- so decisions don’t just get made, they get made faster and with more confidence.

This article isn’t a product pitch. It’s not about ‘how to build a better bar chart’. It’s about something bigger:

How do we make decisions faster- with more clarity, more consistency, and more confidence- using the data we’ve already got?

And how can AI help us do that without overcomplicating it!

If you’re a CDO, Data Director, CFO, or someone in the thick of data strategy- you’ll find practical value here. But if you’re also someone who’s ever stared at a data dashboard and thought, “So what am I supposed to do with this?”– you’re in the right place too.

1. Why Data Dashboards Still Matter (But Don’t Work on Their Own)

Data dashboards alone are no longer enough.

I can hear the gasps and the mental cogs spinning for a response right now.

Yeah, I no longer think they’re enough. Why should they be when there’s now so much more at your finger tips to help than ever before.

The world is moving too fast, and decision-makers are asking for more than a snapshot- they want context, priority, and clarity.

Dashboards still matter. But increasingly, they need help. Help to help a busier (and more impatient) you.

That help is coming from AI.

Dashboards were never meant to make the decision for you. They were built to surface signals, help frame decisions, and nudge people toward action. But when they’re built in isolation- metrics without meaning, trends without translation- they fall short.

AI is becoming less of an add-on and more of a necessity. It spots what’s changed, highlights what matters, and gives us the starting point we often don’t have time to find ourselves.

2. The Real Job of a Data Dashboard: Show Me What to Act On

A good data dashboard doesn’t just show data. It shows consequence.

It answers:

• What changed?
• Why did it change?
• What do we need to do now?

But most dashboards were never built with those questions in mind. They were built to report, not to reveal.

And so we scroll. We hover. We click around.

But the moment AI enters the frame- the conversation inside your head shifts.

Trust me when I say, it can be like a breath of fresh air on that horrid dashboard you have been staring at for the last 20 minutes wondering, blindly, what has changed since the last time you stared at it for 20 minutes.

3. How AI Changes the Game (When the Data Is Right)

I’ve used all sorts of data tools over the years, but the moment AI became embedded into them- something changed. Suddenly, it wasn’t just about visibility, it was about velocity. Speed to spot. Speed to react. Clarity in translating those numbers to words in your head.

Take Microsoft Copilot in Power BI, for example. I’ve explored this first-hand- and while it’s still early days for some organisations, it’s already speeding things up where the groundwork has been done well.

I’ve also spent time playing with tools like ThoughtSpot and Qlik’s Insight Advisor, and while I’m watching closely where they go, the shift is already clear: the assistive layer is no longer optional. It’s becoming expected.

These tools are already helping teams:

• Summarise outliers
• Spot trends in seconds
• Suggest likely next steps
• Flag what needs urgent attention

These aren’t tools of tomorrow. They’re being used today– in meetings, in planning sessions, in everyday decision-making.

But here’s the catch: they only work when the foundation is strong. The principle S*&t in S*&t out, still applies- more so than ever before!

• You need data people actually trust
• Models that reflect how your business really works
• And leadership teams that know what they’re looking to change

AI makes good judgement faster. But it won’t fix vagueness. It can’t generate purpose. That clarity still has to come from you.

4. Designing for Decision Velocity

This is where most data design breaks- and where real opportunity starts.

In any organisation, business leaders are constantly under pressure to make faster, clearer, more accountable decisions. But what feeds those decisions is often fragmented. Dashboards become too dense. Reports too static. And the human in the room? They’re left toggling between gut feel and over-explaining a spreadsheet.

Designing for decision velocity starts with one idea: what does this person need right now to make the call?

For the business, it means:

• Knowing the moment something needs attention
• Seeing the impact quickly
• Getting a next-step suggestion without having to hunt for it

For the data side, it means:

• Building fewer layers between the question and the insight
• Pre-modelling the likely scenarios that affect outcomes
• Embedding real-time summaries, trend callouts, or exception alerts without major work

And for AI, it means:

• Acting like a spotlight, not a crystal ball
• Highlighting what changed, where it matters, and what’s rising up

We don’t need more visualisation. We just need to get to the point quicker- so people can see what matters and move with confidence.

Not just dashboards. Decision engines.

5. What AI Can’t Do (Yet)

AI can:

• Summarise
• Highlight
• Recommend

But it can’t:

• Interpret values
• Judge culture
• Understand context beyond its data…. unless you are already on the way to building our your models.

That’s still the job of leaders. To weigh up what the data says, against what the organisation feels.

Sometimes the data’s telling me one thing, but the room is saying another. That’s when you have to remember- AI doesn’t read the room. You still have to.

But it won’t stay that way forever. AI WILL start getting better at reading the room- not emotionally, but contextually. Over time, with enough examples, decisions, and outcomes logged, it’ll start spotting how we’ve acted before and why. It won’t feel human, but it will start to mimic judgement. And that’s where things get interesting- not because it replaces us, but because it starts anticipating how we lead.

Good data dashboards don’t replace human thinking. They make it faster. AI helps with that. But it still needs the human at the helm.

6. How to Start Rethinking Your Data Design Now

If you’re going to rethink how data supports decision-making, you need to build a method- not just a new dashboard. Here’s a simple structure that works for me in the ‘age of AI’. Also, I coined this, its mine, you cant steal this like I stole a friends analogy in an earlier article ;).

The RETHINK Method:

• R – Reconnect to Purpose: Start with what matters. What is the business or you actually trying to do?
• E – Expose the Questions: What are the questions you or people are struggling to answer today, now?
• T – Test What’s Working: Which dashboards or reports already help, and which slow things down? Scrap anything that slows things down. In fact scrap most and see who screams for a restore (always healthy Data routine this I believe).
• H – Highlight the Gaps: Where is context missing? Where do people ask for more explanation in the room? What are they confused about or left wanting more of?
• I – Integrate AI Thoughtfully: Add AI only where it reduces time to clarity- not where it adds layers. Could simply output a ‘one-liner’… that’s perfect in most cases (remember ‘velocity’).
• N – Note the Outcomes: What decisions are being made? How confident are people in those decisions? **This is new. I’ll come back to this shortly.
• K – Keep the Loop Tight: Make sure learnings from each decision feed back into the system- so over time, the machine knows what worked, what didn’t, and why.

This last part is where AI starts to evolve beyond a helpful assistant.

**If you want AI to really read the room one day- it needs to be trained on the decisions we made, how we got there, and what happened next. But to start this- you need a mechanism of recording decisions in the room wider than ‘minutes’. Start looking at how you structure your Data in your Silver/Gold layers to be able to take in external input such as ‘report decision outcomes‘. Then you can truly train a model.

AI might not feel intuition, but it can learn these patterns from these decisions. It can start to recognise when something just doesn’t smell right– because it’s seen enough history to flag the difference.

Anyway, that’s the long game. That’s leadership through data design. And it starts by closing the loop- not just presenting data, but shaping how decisions are made, captured, and learned from. Closing the Loop. Completing the Loop.

Final Thought: The Real Value Is Clarity

This isn’t just about data dashboards. Or AI.

It’s about leadership- data leadership, personal leadership, room leadership… all the types that contribute towards removing slow paced confusion and spinning up accurate speed.

AI can spotlight what matters, speed up analysis, and surface insights you might have missed. But that doesn’t make your judgement any less vital- it makes it more vital. Because in a world where more is being surfaced, faster, you still need to know what’s real, what’s noise, and what to act on.

AI doesn’t replace your understanding of the data- it sharpens it. If the data is clear, AI helps bring it to life. But if the data’s flawed or the purpose behind it is vague- AI will just accelerate the wrong direction. Biggest warning I can give here.

The future of dashboards and AI is certain- they’re staying. But whether that future serves you depends on how intentionally you lead through it.

AI isn’t the future. But it can be your future- if you choose to use it with purpose.

Thanks again for taking the time to read my articles. Please subscribe to the newsletter to have these sent directly to you.

What’s Next?

There are so many ‘Reflective’ posts/articles out there about AI and Data- I think the feeling is understood (helps, not quite there if you don’t commit, can massively help).

So I’m going to move on.

I’m going to write next weeks articles around one of favourite parts of my CTO role, the creative part, the part that brings the best joy or the worst misery- ‘Product‘.

This will land towards the end of the weekend.

Thanks again 🙂

#BeyondTheTitle #DataLeadership #DecisionDesign #AIDashboards #TechThatLands #DataLeadership #DecisionDesign #AIDashboards