A dashboard is a decision surface. If the page does not change what someone does next, the design pulled its punches.
Start with the decision
Most BI work I have rescued was built backwards. Charts first, audience second, decision somewhere on the way to retirement. The fix is small and unfashionable. Ask one question before the layout opens.
If the answer is vague, you end up with a museum of leftovers. People walk through, nod politely, and act on whatever Slack thread was loudest that morning.
A page worth its scroll knows who reads it, what they can actually move, and which metric carries the weight when the room goes quiet.
One page, one question
The strongest dashboards I have shipped were rude about scope. Are we losing margin to discounts? Which channel pays back qualified pipeline? Where is operational latency turning into churn?
A single page handling three strategic questions and four tactical ones stops being read as a story. It gets scanned for excuses.
Movement, cause, next move
Weak BI pages stop at movement. Revenue up. Conversion down. SLA breaches up again. That is reporting. The story is missing the next two layers.
Decision-ready BI has three: what changed, why, and what we should do about it. Drop the third and the meeting will invent a "what we should do" on the spot, usually with more confidence than data.
Narrative lab
One page, one decision.
Hierarchy talks quieter than colour
A loud dashboard is not the cure for a confused one. Hierarchy is.
The order I keep returning to:
- one headline KPI or takeaway
- one supporting trend or comparison
- one small layer of diagnostic detail
- one named next action with an owner
Narrative stack
A good BI page moves in layers, not in noise.
The reader should reach the answer before the meeting turns into live chart interpretation.
01
Decision question
State the operating question first. If the question is vague, the page will start collecting leftovers.
02
Headline signal
Make the main KPI or takeaway obvious before the diagnostics begin competing for attention.
03
Explaining view
Show the chart that explains the movement. This is where the page earns the right to say why.
04
Diagnostic detail
Keep the supporting breakdown small. Useful detail is fine. A chart museum is not.
05
Next action
Name the owner, the move, or the recommendation. Otherwise the room will improvise one anyway.
Weak example
Busy, colorful, and unhelpful.
Revenue
$1.84M
CTR
4.7%
NPS
61
No single question anchors the page.
Color is used as decoration rather than meaning.
The viewer leaves with movement, but not with a decision.
Strong example
Clear question, visible answer, obvious next step.
Question: which drivers explain the revenue miss enough to change the next operating move?
Headline
-$175K
Most of the miss came from churn and discounting, not from demand collapse.
Next action
Reduce discretionary discounting this week. Owner: growth finance.
The page is about one question, so the hierarchy can stay calm.
Color is semantic, not decorative.
The audience can move from chart to action without an interpretive monologue.
Visual review
Weak and strong choices look different long before the meeting starts.
The point is not taste. It is whether the page makes the reader spend attention on the signal or on recovering basic meaning.
Trend framing
Weak
Too much chrome, weak signal hierarchy, and no clue where the eye should land first.
Strong
One clear trend, one reference, and enough annotation to spare the room a guided tour.
Annotation beats chart volume
Ten unopinionated charts rarely outperform one chart with a useful annotation. A single annotation can name the anomaly, anchor the period, mark the thing as serious or boring, and save you the live narration.
BI storytelling is editorial work in disguise. You are removing ambiguity before someone in the room invents their own.
Executive pages earn trust through boring virtues
Trust in analytics comes from consistent definitions, stable labels, clear date ranges, visible ownership, and the caveats nobody enjoys writing. If a KPI changes meaning between Monday and Friday, the chart design stopped being the main problem some weeks ago.
What I cut first
When a dashboard feels crowded, I take the easy stuff out before I touch real signal. Decorative gradients. Duplicate legends. Unlabelled small multiples. Secondary metrics with no decision attached. Any chart that exists because the data was lying around.
The story usually improves the moment the page becomes slightly less proud of itself.
A review checklist
Before I ship a dashboard page, I ask:
- Can I say the page's question in one sentence?
- Is the main chart the obvious answer to that question?
- Is there a named next action, owner, or recommendation?
- Would a smart stakeholder understand the page without my voiceover?
If the answer is no anywhere, the page is still a draft.
The point of BI storytelling is to make action more obvious than hesitation. That usually means fewer charts, sharper framing, and enough narrative discipline to spare the audience a live archaeological dig.
The chart can still be elegant. It also has to take responsibility for what happens next.