ServerVultr

Why Dashboards Fail (and What Most Organisations Get Wrong). For many organisations, dashboards start with optimism and end in frustration.

Power BI looks promising. Reports get built. Stakeholders are excited. And then, slowly but surely, confidence erodes.

• People stop trusting the numbers.
• Reports multiply.
• Decisions revert to gut instinct.

The problem isn’t the dashboard tool. It’s something far more fundamental.

�

A quick note on dashboards vs reports in Power BI

In Power BI, dashboards and reports are technically different things.

• Reports are multi-page, interactive, analytical assets. They are built from a single semantic model and are used to explore, slice, and drill into data. This is where most analysis and questioning happen.
• Dashboards, by contrast, are single-page, high-level summaries. They are made up of pinned visuals from one or more reports and are designed for monitoring, awareness, and at-a-glance views, often for senior stakeholders.

In practice, however, many organisations use the term “dashboard” as shorthand for all business reporting and analytics, including reports, scorecards, and executive views. That is the sense in which the term is used in this article.

The failure patterns described here apply equally to dashboards and reports: lack of shared definitions, unclear ownership, decision-free design, and cultural assumptions.

Dashboards don’t fail because of technology

Most organisations assume dashboards fail because:

• The data model isn’t quite right
• The visuals could be better
• Performance needs tuning
• Or they need the next platform upgrade

In reality, dashboards fail because they are built before clarity exists. Dashboards are an output. But many organisations treat them as the starting point.

The real reasons dashboards fail

In my opinion, the real reason the dashboards fail is because

1. No shared definition of “the truth”

Ask five people from the same organisation what a key metric means, and you’ll often get five answers. You need a clear definition of

• Revenue
• Active customer
• Conversion
• Engagement

If those definitions aren’t agreed upon up front, dashboards become political tools rather than decision tools.

Without the shared definitions, the result:

• Endless debate
• Manual “sense-checking”
• Offline spreadsheets to “validate” reports

At that point, the dashboard is already dead.

2. Dashboards are built around data, not decisions

A common question during reporting projects is:
“What data do we have?”

The better question is:
“What decisions are we trying to make?”

When dashboards are designed around available data rather than business decisions:

• They become cluttered
• Stakeholders don’t know where to focus
• Important signals get buried in noise

Good dashboards don’t show everything.
They show what matters now.

3. Ownership is unclear

This is often a big one. Who owns the dashboard?

• IT?
• Data?
• Finance?
• The business?

When ownership isn’t explicit:

• Requests pile up
• Changes happen ad hoc
• Nobody is accountable for quality or relevance

Dashboards decay not because people don’t care, but because responsibility is fragmented. I have read this many times in management and leadership books

If three people are responsible for feeding the dog, the dog will starve.

4. Reporting scales faster than understanding

As organisations grow:

• More systems appear
• More metrics get tracked
• More stakeholders want tailored views

Reporting complexity increases exponentially, while understanding does not.

This is why reporting often slows down just when the business needs speed the most.

Dashboards built for a small team rarely survive organisational growth without rethinking structure, governance, and intent.

5. Culture is assumed, not designed

This is where modern platforms, like Microsoft Fabric, are often misunderstood by business decision makers.

Fabric can unify data.
It can centralise tooling.
It can simplify architecture.

But it cannot:

• Align leadership expectations
• Define decision rights
• Create trust in metrics
• Establish accountability

If the data culture is broken, no platform will fix it.

Dashboards fail when clarity comes last

Across organisations of all sizes, the pattern is consistent:

1. Tools are implemented
2. Dashboards are built
3. Questions emerge
4. Trust declines
5. Momentum stalls

The missing step is clarity before construction.

Clarity around:

• What decisions matter
• Which metrics drive those decisions
• Who owns them
• How they are interpreted and challenged

Without that, dashboards become expensive wallpaper.

What successful organisations do differently

From my experience, the organisations that get lasting value from dashboards take a different approach:

• They start with decision-making, not visualisation
• They align leaders before building reports
• They treat dashboards as evolving products, not one-off deliverables
• They fix structure and ownership before scaling tooling

Dashboards then become accelerators, not bottlenecks.

Why this matters

If your organisation:

• Has lots of dashboards but little confidence
• Is “looking at Fabric” but unsure where to start
• Feels reporting should be delivering more value by now

Then the issue probably isn’t technical.

It’s strategic.

And until that’s addressed, dashboards will continue to disappoint—no matter how modern the platform.

Where this fits in the bigger picture

This is the first in a short series exploring why data initiatives stall as organisations grow, including:

These are exactly the problems we see organisations bring into our Data & Analytics Accelerator,long before they ask about dashboards or tooling.

Next steps

If this resonates with you, the next question to ask isn’t:
“Which dashboard should we build next?”

It’s:
“Do we actually have clarity on what we’re trying to achieve?”

That’s where progress starts.

How My Hail Mary Became a Fail Mary. Yet Somehow I Still Won the Gameweek. A fail Mary that works!

As regular viewers of the FPL Power BI SQL Show and long-time readers of this blog will know, we spend a ridiculous amount of time every week talking strategy. Proper strategy. Not gut-feel punts, but decisions driven by the mountain of data we pull from the FPL API. We digest it, interpret it, shape it, turn it inside out and upside down, all in the hope of turning our respective FPL teams into league winners rather than the mid-table fodder they sometimes like to pretend they are.

And last week’s data was interesting. Very interesting. Looking at fixture difficulty, predicted scores, and the historic performances of certain high-end assets, it felt like the door was cracking open for what I like to call a “bit of play.” And by “bit of play,” I of course mean: this might be the week I try something stupidly brave and hope it pays off. Or, more politely, an opportunity to differentiate my strategy and claw back points on the rivals who have been getting a bit too comfortable above me.

Spotting the Opening: Haaland’s Tough Fixtures

Going into the gameweek, the numbers suggested Haaland was about to enter a tricky spell, away at Liverpool and away at Newcastle either side of the international break. Historically, he hasn’t exactly taken Liverpool to the cleaners, and with that extremely high effective ownership, you know what that means: sometimes the biggest differential is simply not captaining the player everyone else slaps the armband on without thinking.

I had been mulling this over for a while. The thought crossed my mind earlier in the week: If I’m feeling brave… and if the stars align… maybe I take the armband off Haaland for the next two gameweeks and hope for blanks. Not selling him. Not benching him. Just quietly removing that shiny captaincy badge and praying to the FPL gods.

I even mentioned it on the FPL Show. I called it my Hail Mary moment, you know, the one where I make a play that either wins me the gameweek or tanks me beyond recovery. I floated the idea to Justin, who gave me the classic Justin response: “Errr… I dunno,” combined with that face that says he thinks I’ve completely lost the plot. He also mentioned he was quitting Meta, and we need to move our comms to Telegram or Signal — but I’ll let Justin explain that particular life decision.

The Play: Armband to Gabriel

Anyway, I made the play. I took the armband off Haaland… and I gave it to Gabriel at Arsenal.

Now, to be fair, this wasn’t random. Gabriel had the highest predicted score in the dataset that week. Arsenal hadn’t conceded in eight or nine games. And with their “Crazy Gang” style of lump-it-up-and-head-it-in football, he’s always lurking for a goal. Arsenal were playing Sunderland — newly promoted, decent start, but not exactly a side the so-called “title favourites” should be struggling with. If Arsenal really are as good as half the pundits insist they are, this should have been a straightforward put-them-to-the-sword job.

But Arsenal… oh Arsenal. They drew 2–2 with Sunderland. At home. Proving to themselves and the world that perhaps the title trophy isn’t being engraved with their name in November after all, and maybe — just maybe — we need to see how the other 27 games pan out before crowning the “New Crazy Gang” as champions-elect.

Gabriel, in the middle of that absolute circus of a performance, managed a mighty one point. So with the armband, I got two. Glorious.

Bird, of course, immediately messaged me with the perfect summary:

Fail Mary indeed.

All Was Not Lost… Somehow

Here’s the funny thing though: aside from that disgrace of an Arsenal showing, the rest of my squad was actually ticking along pretty nicely. My Crystal Palace and Chelsea assets pulled their weight. My squad players did what they were meant to do. I’d gone into the week deciding not to burn transfers on panic moves and instead wanted to see what my bench could deliver. And to be fair to them — they delivered.

Still, the entire strategy hinged on one thing: Haaland needed to blank.

Fast forward to the final match of the weekend — Man City vs Liverpool. City absolutely destroyed Liverpool. The only reason the scoreline wasn’t uglier was a dodgy decision to disallow Liverpool’s goal. City were dominant. And Haaland scored.

Disaster?

Well… sort of. But not entirely. Because Haaland also missed a penalty before he scored. That dragged his total down. Captainers got 8 points. I got 4. Net loss: 4 points.

Painful? Yes.

Season-ending? Nowhere near.

With the rest of my team performing respectably, it turned out that my Fail Mary… somehow wasn’t a full fail after all.

The Final Tally: 51 Points and a Chunk Off Justin

I ended the gameweek on 51 points. On paper, that’s not exactly inspiring. It’s the sort of score that feels like someone’s left the handbrake on.

But here’s the bit that matters:

It was 14 points above the gameweek average.

And more importantly…

It closed the gap on everyone around me — including taking a lovely, satisfying chunk out of Justin’s lead.

So despite producing one of the biggest captain fails of the week, I still won the gameweek in terms of rank movement and rival damage. A reverse Fail Mary. A Fail Mary that works. A statistical miracle delivered courtesy of missed penalties, over-performing bench players, and Arsenal being absolutely incapable of beating Sunderland.

This is exactly why we love (and hate) FPL.

Looking Ahead

FPL is back next week. The data machine is already warming up. I’ve got the models running, the dashboards updating, and the FPL API coughing out numbers that will, once again, tempt me into doing something bold, stupid, genius, or all three at once.

I’ll keep you posted — especially if I decide to perform another Hail Mary. Or Fail Mary. Or whatever this strange hybrid version was.

Because if there’s one thing I’ve learned this season:

Sometimes doing the “wrong” thing is the only way to make the right kind of progress.

And sometimes… the Fail Mary works.

Useful Links

What is a data strategy?

Why Every SME Needs a Data Strategy (Even If You Think You’re Too Small)

The Cost of Doing Nothing: How Ignoring Data Strategy Drains SME Growth

�