Dashboard Theater: Why Your Board Metrics Might Be Lying to You

I've sat in a lot of meetings where a fifty-slide deck shows up, every chart green, every trend line pointing the right way. Then someone asks a simple, specific question — say, "what's our exact exposure to this shift, across both our retail and private-wealth books?" — and the room goes quiet. The real answer needs three weeks of manual reconciliation and a few arguments about whose numbers are right.

I call this dashboard theater. The metrics look rigorous. They don't actually hold up.

Why it happens

It's rarely a tooling problem. Most places I've worked have decent BI stacks and smart data teams. The issue is structural: reporting teams build whatever the business asks for that week, one request at a time. A few years of that and you end up with hundreds of dashboards, each one a narrow slice of the truth, none of them talking to each other.

Three things tend to go wrong once you're relying on that kind of patchwork:

• Nobody can trace where a number came from. If risk, wealth, and ops each report "customer acquisition cost" differently, you don't actually have one number — you have three, and no one's sure which is right.
• Everything's late. Decisions wait on end-of-month stitching instead of running on something closer to real time.
• The wrong things get measured. Teams report what's easy to count (total accounts) instead of what actually matters (how fast a new client gets their first allocation).

What's worked, in my experience

The fix isn't a better dashboard. It's treating your key metrics like a product, not a report. A few things that have made a real difference on the systems I've built:

Someone owns each number

Every metric that shows up in front of leadership should have one accountable owner — someone who can explain exactly how it's calculated and stand behind it, not just whoever happened to build the query.

You can trace it back

If someone clicks into a number, they should be able to see where it came from, down to the source system. Especially in regulated industries like banking and wealth management, "trust me" isn't good enough — auditability has to be built in, not bolted on after the fact.

It ties to an actual decision

Before a metric earns a spot on an executive dashboard, it's worth asking: if this number moved 15%, what would we actually do differently? If the honest answer is "nothing," it doesn't belong there.

Fast and mostly right beats slow and perfect

An 80%-accurate number today is often more useful than a perfect one three weeks from now, especially when a decision is time-sensitive. I've seen teams over-invest in precision on numbers that needed to just be timely.

Four questions worth asking

If you want a quick read on whether your own reporting has this problem, these four tend to surface it fast:

1. If you asked three different teams for the same core number, would they give you the same answer without a week of prep first?
2. How many days pass between a period closing and someone actually being able to act on the numbers?
3. How many dashboards got retired last quarter? (If the answer's zero, that's usually not a good sign — healthy systems prune old reports.)
4. What's one real decision made in the last 90 days that was directly driven by one of these reports?

None of this is complicated. It's mostly discipline — deciding what actually deserves a spot in front of leadership, and being honest about the gap between metrics that look good and metrics you can actually act on.

← back to Blog