Omni-Channel Contact Center Dashboard
One operational view across calls, chats, WhatsApp and campaigns — KPIs, channel performance, filters and drill-downs for supervisors.
Client
Enterprise Contact Center Platform (confidential)
Role
UX/UI Designer — information architecture, data hierarchy, dashboard and component library.
Platform
Web Dashboard
Confidentiality: Client identity and operational metrics are confidential. The interface patterns shown are sanitised.
01Overview
An operations dashboard for contact-centre supervisors: calls, chats, WhatsApp and campaigns in one view, with KPIs, channel performance, trends, filters and drill-downs — designed as a decision tool, not a poster.
02Context
Operations teams ran reports in separate tools per channel. Staffing, campaign and coverage decisions were slow because the data lived apart. The dashboard's job was to make one question answerable in seconds: what is happening across our channels right now?
03User Journey
How supervisors work with it
04UX Decisions
01KPI cards, not number walls
Four to six KPIs at the top, each a card with a delta and a sparkline — scannable in a glance, clickable for depth.
02Channels as first-class tabs
Calls, chats, WhatsApp and campaigns get parallel views with identical layout rules, so learning one teaches all.
03Filters with defaults, not empty states
The filter bar opens with a sensible default range, so the page is never a blank void before configuration.
04Drill-down as the only click
One click from KPI to detail; the detail view answers the question the KPI raised.
05Real-time is marked, not animated
Live data is labelled with a refresh timestamp — no gratuitous motion pretending to be live.
05Interaction Design
States, transitions and behaviour — not just the happy path.
Loading
Skeleton cards preserve layout so the page never jumps; refresh states are visible.
Empty state
A named, explained empty state per view ('No calls in this range') with a reset-filters action.
Metric API failure
The affected panel shows a retry state, not a broken page — partial failure is contained.
Stale data
A timestamp plus a 'data is X minutes old' marker keeps stale data honest.
Permission-limited views
Restricted metrics render as explained placeholders, not mysterious blanks.
06Edge Cases
Good UX is what happens when things go wrong. Each case: the problem, the user-facing message and the recovery action.
No data for period
Edge caseA channel has no activity in the selected range.
“No calls in this period.”
Recovery: Empty state with a reset-filter action and a link to a wider range.
API failure
Edge caseOne metric service is down.
“Live metrics unavailable.”
Recovery: Panel-level retry with last-known values shown, page otherwise intact.
Stale data
Edge caseThe data feed lags.
“Data is 15 minutes old.”
Recovery: Timestamp marker plus a refresh control — no silent staleness.
Permission limits
Edge caseThe user cannot see a metric.
“This metric requires supervisor access.”
Recovery: Explained placeholder instead of a confusing blank.
07Final Experience
Omni-channel overview
Omni-channel overview
KPI cards, channel tabs and trends — the decision surface, not a poster.
Schematic placeholder — add an approved screenshot via data/case-studies.ts
Channel detail
Channel detail
Queue, agents, conversations and reasons — one click from the KPI.
Schematic placeholder — add an approved screenshot via data/case-studies.ts
Filter bar
Filter bar
Sensible defaults; the page is never a blank void before configuration.
Schematic placeholder — add an approved screenshot via data/case-studies.ts
Drill-down modal
Drill-down modal
Detail answers the question the KPI raised, then closes without losing context.
Schematic placeholder — add an approved screenshot via data/case-studies.ts
08Learnings
- Dashboards are tools, not posters: define the decision each view supports before designing a single chart.
- Data density is a design problem — hierarchy and spacing decide what is readable.
- Partial failure states (one panel down) are more common than total outages; design for them.
Next case study
AI Agent Assist Dashboard
← /work · case study · omni-channel-dashboard