The problem with most executive dashboards

Dashboards usually start with good intentions and then grow. Every department adds its favorite numbers until the page has 40 charts and nobody knows where to look. The result is a dashboard people open, scroll and close, without changing a single decision.

The fix isn't more data. It's fewer, better metrics, each tied to a question leadership actually needs answered every week.

Weekly Leadership Dashboard

Illustrative example · Sample numbers

Are we selling?

Revenue booked this week

$48.2K

▲ Up vs. 4-week average

Are we selling?

Qualified pipeline

$126K

▲ 3 new opportunities

Are we delivering?

On-time delivery

92%

▼ Below 95% target

Are we getting paid?

Cash on hand

11 wks

● Steady, within target

Are clients happy?

Repeat and referral rate

38%

▲ Up vs. last quarter

Is the team OK?

Team capacity used

87%

▲ Near overload, review hiring

Six numbers, five questions. Each tile shows the value, how it compares to a target or trend, and the last seven weeks.

The five questions every weekly dashboard should answer

Instead of starting with the data you have, start with the questions leadership needs answered. For most growing businesses, they come down to five.

QuestionMetrics to trackRed flag to watch for
Are we selling?New revenue booked, qualified pipeline, inquiry-to-client conversion ratePipeline shrinking for several weeks in a row
Are we delivering?On-time delivery, jobs or projects completed, turnaround timeTurnaround time creeping up while volume stays flat
Are we getting paid?Cash on hand, invoices sent, overdue receivablesOverdue invoices growing faster than revenue
Are clients happy?Repeat or referral rate, client feedback score, complaints or refundsA rise in refunds, cancellations or complaints
Is the team OK?Capacity used, overtime hours, open rolesCapacity above 85 to 90% for weeks at a time

Balance leading and lagging metrics

Lagging metrics tell you what already happened, like last week's revenue. Leading metrics tell you what's likely to happen next, like the size of your pipeline or how many inquiries came in. A useful weekly dashboard includes both, so you can act before the lagging numbers turn bad.

Leading: what's coming

New inquiries · Qualified pipeline · Proposals sent · Bookings for next month · Team capacity

Lagging: what happened

Revenue · Profit margin · Cash collected · Projects completed · Client retention

Leading metrics give you time to act. Lagging metrics confirm whether it worked.

How many metrics is too many?

For a weekly executive view, aim for eight to twelve metrics at most. If a number hasn't changed a decision in the last three months, move it to a detailed report and take it off the main dashboard. Every metric that stays should have three things:

  • An owner who can explain it and act on it.
  • A target so everyone knows what good looks like.
  • A trend so you can see direction, not just a single snapshot.

How to run a 15-minute weekly review

  1. Scan for red flags (3 minutes). Look only at metrics that missed their target or changed direction.
  2. Ask the owner why (7 minutes). Each flagged metric's owner explains what happened.
  3. Decide one action per flag (5 minutes). Assign an owner and a due date, and check on it next week.

When the dashboard updates automatically, this meeting stops being a report-out and becomes a decision-making session.

Frequently asked questions

What's the difference between a KPI and a metric?

A metric is any number you track. A KPI is one of the few metrics tied directly to your most important goals. Your weekly dashboard should mostly show KPIs.

Should the dashboard update automatically?

Yes, whenever possible. If someone has to rebuild the dashboard by hand every week, it will eventually fall behind, and leaders will stop trusting it.

What tools can I use to build a weekly dashboard?

Power BI, Looker Studio and even a well-structured Excel file connected to your data sources can all work. The right choice depends on where your data lives and what your team already uses.

Keep reading