The Dashboard Had the Numbers — But Not the Answers
Redesigning reporting around business questions instead of metric lists, creating a dashboard that drives decisions rather than displaying numbers.
The Result at a Glance:
Dashboard had 40+ metrics but no clear answers
Reporting organized around available data, not business questions
Restructured around 5 core KPIs tied to 3 business questions
Stakeholders moved from reviewing data to making decisions. Dashboard adoption increased and manual reporting time decreased significantly.
Analytics / Data Visualization
Stakeholder interviews → KPI definition → dashboard architecture → build → insights
Stakeholder Context
The existing dashboard suite contained over 40 metrics across multiple pages. Stakeholders opened it weekly but could not consistently identify what was improving, declining, or requiring action. The dashboard was comprehensive but not actionable.
Business Questions
I interviewed stakeholders to identify the actual questions they needed answered:
- Are conversions declining, and if so, where in the funnel?
- Which channels are driving qualified revenue vs. low-quality traffic?
- Where is the funnel weakening this week compared to last week?
KPI Definition
I created a KPI dictionary defining exactly how each metric is calculated, what data source it uses, and what constitutes a meaningful change:
| KPI | Definition | Source | Review Cadence |
|---|---|---|---|
| Conversion Rate | Purchases / Sessions | GA4 | Weekly |
| Revenue | Sum of purchase_revenue | GA4 + Shopify | Daily |
| Blended ROAS | Revenue / Total Ad Spend | Ads + GA4 | Weekly |
| CAC | Ad Spend / New Customers | Ads + CRM | Monthly |
| Funnel Drop-off | Step N / Step N-1 | GA4 | Weekly |
Dashboard Architecture
I structured the dashboard in layers of detail:
Analytical Findings
The restructured dashboard immediately revealed patterns that were invisible in the old format. Conversion trends were visible by segment, not just in aggregate. Channel performance was comparable side-by-side. Funnel weaknesses were identifiable by step and by device.
Recommendations
Based on the dashboard redesign process, I recommended:
- Start every dashboard with the 3 questions it answers — not the data it contains
- Define KPIs before building any visualization
- Use consistent date ranges and comparison periods across all pages
- Add written commentary to weekly reports — charts alone don't explain "so what?"
Before / After
The transformation reduced cognitive load while increasing analytical depth. Stakeholders spent less time searching for information and more time acting on insights.
| Dimension | Before | After |
|---|---|---|
| Metrics displayed | 40+ | 5 core KPIs + drill-down |
| Time to find answer | Minutes of searching | Visible on first page |
| Decision clarity | Ambiguous | Clear action signals |
| Dashboard adoption | Opened but rarely used | Referenced in weekly meetings |
Outcome
The redesigned dashboard became the single reference point for weekly performance reviews. Stakeholders transitioned from reviewing numbers to making decisions.