Designing a forecasting tool for call centre resource planning
My role
Visual Designer
Year
2020
Project details
Client
Vodafone
Sector
Telecommunications
Project team
Small product team including a Project Manager and a UX Designer.
Timeline
2 weeks
Overview
I designed the user interface for a Vodafone call-centre optimisation tool, helping managers compare live performance against targets and explore staffing scenarios before making operational decisions.
The product was designed for a data-dense environment where managers needed to understand queue health, performance risk and resource options quickly. The challenge was to turn complex operational wireframes into a clear, accessible and production-ready experience across desktop and mobile.
My role focused on visual design, interaction detail, design-system extension and developer handoff. I worked from end-to-end UX wireframes, applying Vodafone’s design system and creating additional patterns for high-density data, forecasting controls and status-heavy interfaces.
The problem
Call-centre managers needed a clearer way to understand operational performance and make fast resource decisions.
Existing workflows relied on a mix of dashboards, spreadsheets and operational tools, making it harder to see where queues were at risk, how performance compared with targets, and what staffing adjustments might help.
The product needed to support managers who were asking practical questions such as:
Which queues are at risk?
Where are we above or below target?
What happens if agents are moved between queues?
Which changes are most likely to improve service levels?
What needs attention now?
The design challenge was to make this information readable and actionable without oversimplifying the operational detail managers needed.
My role
I worked as the visual designer on the project, translating UX wireframes into high-fidelity screens and production-ready specifications.
My work included:
reviewing wireframes, requirements and operational use cases
defining visual hierarchy for dense performance data
applying Vodafone’s design system across desktop and mobile layouts
creating UI patterns for KPIs, variance, queue health and forecasting controls
designing table, status, alert and drill-down states
supporting accessibility through colour-independent status cues and readable data layouts
producing interactive prototypes
preparing developer-ready specifications covering layout, states and behaviours
documenting reusable patterns for future design-system use
The project required speed, precision and pragmatic design decisions because the delivery timeline was only two weeks.
Design approach
The main focus was to help managers move from performance data to operational decisions more quickly.
Rather than creating a dashboard that showed everything with equal weight, the interface needed to prioritise risk, variance and action.
The design approach focused on three priorities:
glanceability; so making current performance, target variance and queue risk easy to scan
decision support; so helping managers understand the potential impact of staffing or routing changes
implementation clarity; so creating detailed UI specifications that could be built consistently and reused across the product
Key design decisions
Make current versus target performance obvious
Managers needed to understand performance gaps quickly. I designed KPI and status patterns that compared current performance against target, with clear variance indicators and trend signals. The goal was to help users spot risk without needing to interpret raw numbers across multiple sources.
Prioritise queue health and operational risk
In a call-centre environment, not every data point carries the same urgency. I used hierarchy, grouping and status patterns to make at-risk queues more visible. This helped the interface direct attention towards areas where action may be needed, rather than presenting all queues as equally important.
Support “what-if” scenario thinking
Managers often needed to consider possible staffing changes before committing to them. I designed forecasting controls that allowed users to adjust variables such as time window, queue thresholds or agent allocation and preview projected impact. This helped turn the tool from a passive dashboard into a decision-support interface.
Keep actions close to insight
Operational tools become slower when users have to move between separate screens to understand a problem and take action. I placed actions such as reassignment, rerouting and scheduling prompts close to the data they affected. This reduced unnecessary context switching and kept decision-making connected to the relevant queue or performance signal.
Use progressive disclosure for dense information
The product needed to contain detailed queue, skill and agent-level information without overwhelming the main view. I designed overview states that surfaced critical signals first, with drill-downs for users who needed more detail. This helped keep the main interface readable while preserving access to operational depth.
Extend the design system for data-heavy use cases
Vodafone’s design system provided the foundation, but the product needed additional patterns for dense operational data. I created or adapted components for high-density tables, status chips, KPI cards, warning states and forecasting controls. These patterns helped create consistency across the product and gave engineering clearer reusable elements to build from.
Designing for accessibility and clarity
Because the interface relied heavily on status, alerts and performance thresholds, accessibility was an important design consideration.
I avoided relying on colour alone to communicate status. Key states were supported with labels, icons and clear hierarchy so that warning, breach and neutral states remained understandable in dense layouts.
I also paid attention to typography, spacing and table density to keep information readable in an enterprise context where users may be scanning large amounts of data under time pressure.
Testing and iteration
The project moved quickly, so feedback came mainly through stakeholder reviews, operational walkthroughs and feasibility checks with engineering.
Iteration focused on:
improving readability of dense data layouts
clarifying hierarchy between KPIs, queues and actions
refining status and threshold patterns
checking feasibility of forecasting interactions
aligning components with Vodafone’s existing design system
reducing ambiguity in handoff documentation
The work was shaped by the need to move fast while still producing a reliable interface for implementation.
Outcome
The project delivered high-fidelity desktop and mobile prototypes for a call-centre optimisation tool, along with developer-ready specifications and reusable UI patterns.
The work translated complex operational wireframes into a clearer, more structured interface for reviewing performance, identifying risk and exploring resource scenarios.
I do not have post-launch operational metrics for this work, so I avoid overstating the impact. The clearest outcome I can point to is that the project produced a production-ready visual design direction, extended the design system for data-heavy operational use cases, and gave the delivery team clearer specifications for implementation.
What I learned
This project reinforced how important visual hierarchy is in data-dense operational tools.
When users are making fast decisions, the interface cannot simply display more information. It needs to make risk, priority and action easier to recognise.
It also showed the value of designing reusable patterns for complex enterprise interfaces. Components such as KPI cards, status chips, tables and forecasting controls need to work together as a system, especially when the product has to scale beyond a single screen.
Reflection
I chose this project because it shows my experience designing data-heavy operational interfaces where clarity, density and implementation quality all matter.
It required visual design craft, accessibility consideration, design-system thinking and the ability to turn complex wireframes into a polished product experience within a very short delivery window.
Designing a headless CMS for scalable digital content operations
Led UX for a 13-person product team, helping IKEA move from manual specification documents to a structured workflow for creating and publishing shoppable content at scale.
