Ecommerce Dashboard Design: What Should Ecommerce Teams Measure?
How to design ecommerce dashboards: information architecture, hierarchy, audiences, drill-downs, comparisons, chart choice, alerts, accessibility and governance.
Quick answer
Good ecommerce dashboard design starts with a question and an audience, not a list of KPIs. Put the question at the top, show a few headline metrics against targets and prior periods, explain movement with drivers (traffic, conversion, order value) and breakdowns (channel, device, category, product), provide drill-downs to funnel, product, customer and operations views, annotate events, surface alerts sparingly, use simple accessible charts and build everything on agreed definitions. Design separate dashboards for leadership, merchandising, marketing, CRO and operations.
Why Dashboards Fail
Most ecommerce dashboards fail through clutter, not missing data: dozens of equally sized tiles, no comparisons, unclear definitions and no path from “revenue is down” to “why”. People stop opening them. Dashboard design is information architecture: deciding what matters most for whom, in what order and at what level of detail. For choosing the metrics themselves, see ecommerce KPI dashboard.
Start With Questions and Audiences
| Audience | Core question | Typical content |
|---|---|---|
| Leadership | Are we on track? | Revenue, orders, conversion, AOV, margin vs target |
| Merchandising | Which products and categories need attention? | Product and category performance, stock, returns |
| Marketing | Which channels and campaigns work? | Traffic, conversion and revenue by channel, spend efficiency |
| CRO / UX | Where do shoppers struggle? | Funnel steps, device, search, filters, experiments |
| Operations | Are we fulfilling well? | Orders to ship, delivery times, returns, stock issues |
| Customer / CRM | Are customers coming back? | Repeat rate, cohorts, subscriptions, churn |
Information Hierarchy
Structure each dashboard in three levels. Headline: the few numbers that answer the question, with comparisons. Drivers: the components that explain headline movement (for revenue, that's sessions × conversion × average order value, or new vs returning customers). Diagnostics: breakdowns and lists that point to specific causes (channel, device, category, product, market). The mock-up above follows this structure, ending with drill-downs into detailed views.
Revenue = Sessions × Conversion rate × Average order value
Revenue down 8% vs last week
Sessions +2% (paid social up, organic flat)
Conversion −9% (mobile −15%, desktop −1%)
AOV −1%
→ drill into mobile funnel: checkout payment step −20% after release on TuesdayComparisons and Context
A number without context can't be interpreted. Show each headline metric against a target and a comparison period (previous period, and same period last year where seasonality matters), label time windows clearly and keep them consistent across tiles. Annotate events that explain changes: campaigns, releases, outages, stock issues, pricing changes. See ecommerce analytics.
Dashboards nobody opens?
ZSpace designs ecommerce dashboards around the questions each team asks, with clear hierarchy and drill-downs.
Choosing Visualizations
| Need | Visual | Avoid |
|---|---|---|
| Headline value and change | Number tile with delta and sparkline | Gauges and dials |
| Trend over time | Line chart with annotations | 3D charts |
| Compare categories or channels | Sorted bar chart | Pie charts with many slices |
| Product performance detail | Sortable table with the full path | Dozens of small charts |
| Funnel | Step bars with rates | Decorative funnel shapes without numbers |
Drill-Downs and Navigation
Every headline should lead somewhere: revenue to channel and category views, conversion to the funnel by device, product performance to product detail. Keep filters (date range, market, device, channel) consistent across views, and preserve them when drilling down. See funnel analytics and product analytics.
Alerts and Data Health
Alerts draw attention to what changed significantly: conversion drops on a device, a product selling out, tracking gaps. Keep them few, explain the threshold and link to the relevant view. Show data freshness and known tracking issues on the dashboard so people don't act on broken data. See analytics architecture.
Accessibility and Readability
- Sufficient contrast for text and chart elements
- Good and bad not shown by red and green alone; add arrows or labels
- Direct labels on charts where possible
- Tables or text summaries alongside charts
- Consistent number formats, currencies and units
- Keyboard-accessible filters in interactive tools
Governance
Dashboards are only trusted when definitions are shared. Link each metric to its definition, record owners, review dashboards regularly and retire unused ones. Building dashboards from a modelled data layer rather than raw events keeps numbers consistent across teams.
Worked Example
An illustrative scenario: an ecommerce team's main dashboard has 40 tiles, no targets and conflicting conversion figures. The redesign creates a leadership dashboard (revenue, orders, conversion, AOV and margin vs target and last year, with driver breakdown and annotations), a merchandising dashboard (category and product tables with the full path and returns) and a CRO dashboard (funnel by device, search and filter metrics, experiments), all from shared definitions with drill-downs. Usage of each dashboard is tracked.
A Dashboard Design Process
| Step | Output |
|---|---|
| 1. Interview users | Questions, decisions and frequency |
| 2. Define metrics | Definitions linked to the data model |
| 3. Sketch hierarchy | Headline, drivers, diagnostics, drill-downs |
| 4. Prototype | Low-fidelity layout tested with users |
| 5. Build on modelled data | Consistent numbers |
| 6. Review usage | Retire or improve unused views |
Common Mistakes
- Listing every available metric
- No targets or comparisons
- Charts chosen for decoration
- Different definitions in different dashboards
- No drill-down from headline to cause
- Colour-only status indicators
Ready to design dashboards your team uses?
Talk to ZSpace about dashboard UX and information architecture, analytics audits and automated reporting.
Conclusion
Ecommerce dashboard design is information architecture for decisions: questions and audiences first, then headline, drivers and diagnostics, comparisons, drill-downs, accessible visuals and shared definitions. For the metrics themselves, see ecommerce KPI dashboard.
Common questions
It answers specific questions for a specific audience, leads with a few headline metrics against targets or comparisons, explains changes through drivers and breakdowns, offers drill-downs for detail and uses clear, accessible visuals with agreed definitions.