The dashboard graveyard
I have built dashboards for a decade, and I have also buried a few - beautiful, complete, and unvisited. The pattern behind every dead dashboard is the same: it answered every possible question, so it answered nobody's actual question.
What survives: three numbers and coffee
The dashboards that get opened every morning have a different shape. Three numbers, big, with yesterday next to them. For an e-shop: revenue yesterday, orders in progress, ad spend against it. For a services firm: cash position, pipeline value, hours billed this week. The owner checks them with coffee, and when one looks wrong, THEN they click deeper.
Fewer numbers is not a limitation - it is the feature. A number earns its place on the morning screen only if a bad value would change what you do today. Everything else belongs one click down.
Finding your three
The exercise I run with owners takes twenty minutes: imagine you are at the airport, you have one glance at your phone before boarding. What three values tell you whether to relax or to make a call? The first two come instantly. The third is where the argument starts - and that argument is the most valuable part, because it forces the question "what do we actually steer by?"
The plumbing matters more than the charts
A dashboard is only as honest as its pipes. If the stock number comes from the file someone updates "usually on Fridays", you have built a very fast way to look at stale data. Most of my dashboard work is actually integration work - pulling the truth out of the e-shop, the accounting system and the warehouse before drawing anything. The chart library is the last and least interesting decision.
If your reporting means someone assembles a file every Monday, tell me what is in it - that file is the spec for a dashboard that builds itself every night.