Designing Real-Time Intraday Adherence Monitoring

Story at a glance

Before—>

After—>

Challenges and my role

A SaaS workforce management product I worked on, was covering the full WFM lifecycle. At the time of this project, intraday management was in discovery and lacked customer-facing features, despite strong foundational capabilities already in place.

Preconditions

• Core WFM products (scheduling, forecasting, time tracking) were live and stable.
• Required data sources were available:
  • Real-time agent status from third-party systems via integrations.
  • Agent status generated internally through the time-tracking product, related case study

Challenges

Business perspective

The platform lacked a real-time adherence monitoring solution, limiting team lead’s ability to manage schedule compliance throughout the day. This gap impacted:

1. Customer retention, due to an incomplete WFM offering
2. Product adoption, especially for intraday use cases
3. Customer satisfaction, as leads lacked timely operational visibility

Product perspective

The product needed to enable team leads to:

1. Monitor agent adherence in real time
2. Identify trends and patterns at agent and team level
3. Act quickly on schedule deviations to minimize service-level impact

My role

As the Product Designer, I was responsible for:

• Conducting customer interviews to uncover needs and pain points
• Owning the end-to-end design process (research, ideation, prototyping, validation)
• Collaborating closely with PMs and engineers to align design and delivery
• Ensuring solutions aligned with and scaled through the existing design system

Understanding users

Team lead

Also known as “Real Time Manager”, depends on the company structure and size.

Agent

What is real matters

Real life of agent

Why it is bad

• Missing customers’ calls.
• Increase customer waiting time.
• Decrease customer satisfaction.
• Lost deals.

Approach solution

Problem and opportunities

I followed a continuous discovery approach, inspired by Continuous Discovery Habits by Teresa Torres, to understand user goals and validate assumptions early.

Interivew users acround core assumptions

I created an interview guide that:

• Explicitly stated goals and assumptions
• Focused on open-ended questions
• Clearly separated research questions from interview prompts
• This ensured conversations surfaced real behaviors rather than opinions.

Synthesizing interviews insights

After each interview, I summarized key insights and reviewed them with the Product Manager to validate or challenge our initial assumptions.

Define opportunities and feasibility

I used an Opportunity Solution Tree to connect desired outcomes with potential solutions.

I then applied the Kano model to prioritize opportunities based on user value and implementation effort.

Framing of trade-offs

Problem statement

Team leads needed to detect schedule deviations early enough to act, but showing all agents current and schedules activity table will overwhelmed them with real-time data, making it difficult to distinguish actionable anomalies from background noise.

Strategy

Design a real-time adherence dashboard that surfaces significant deviations first, enabling team leads to identify and address issues quickly without cognitive overload.

Design Priority

We prioritized fast anomaly detection and clarity over exhaustive data exposure. To support rapid scanning and interpretation:

• Out-of-adherence reasons were grouped into three meaningful categories
• Visual hierarchy emphasized severity and deviation impact
• The primary view focused on exceptions, not full activity listings

Out-of-adherence events contribute to the adherence score only when the duration exceeds a configurable threshold (default: 3 minutes). This allows team leads to filter out short-term noise while retaining control over sensitivity.

Decision rationale

Why this path

The dashboard is the primary surface for monitoring real-time adherence, so I explored multiple design directions before converging on the final solution.

• Compare the current and scheduled activity side-by-side
• Highlight out-of-adherence agents clearly, enabling quick identification of issues
• Differentiate severity visually: agents reach the threshold are shown with a lighter visual treatment.

Alternatives considered

I explored more minimal table-based layouts that reduced contextual information and relied primarily on activity data.

Why they were not selected:

• Reduced context increased cognitive effort when identifying and acting on deviations
• Repeated exception codes made legends unnecessary and harder to interpret

Design outcome

Real-time dashboard

Team leads can monitor real-time intraday performance, also adjut the scheduled activity if needed.

Historical adherence

Agent intraday adherence

Using real-time adherence data, performance is aggregated per agent to provide a detailed activity history. The monthly view helps team leads:

• Identify behavioral trends and recurring deviations
• Compare high and low performers
• Support coaching and performance reviews with concrete data

Team intraday insight

Team leads can assess overall team adherence performance and spot emerging patterns across the day.

• This supports operational decisions such as:
• Detecting overload situations (e.g. sustained high call volume)
• Identifying systemic schedule deviations

Constrains

• Agent status data imported from third-party systems required additional mapping due to legacy code
• A mobile version was not delivered due to the density of data and interaction complexity
• Without a direct connection to staff requirement data, the product could not provide automated rescheduling recommendations.

Impact and reflection

Customer feedback

Usages statistics

Over the past 12 month (2024 Septmber to 2025 August), used by about 100 tenants:

• Historical intraday: ~300 interactions per day over the last 12 months.
• Real-time adherence: ~80 interactions per day over the last 12 months.

Interpreting usage patterns

• Historical intraday (most used):

  • Primary use for payroll validation and end-of-day performance reviews
  • Secondary use for weekly and monthly reporting to management

• Real-time adherence (less used):

  • Primarily used for monitoring rather than interaction
  • Adoption constrained by regional regulations (e.g. strict employee monitoring policies in Germany)

Reflection

Lessons learned

• Team-level historical adherence had limited value: team leads needed insight into specific deviations, not high-level status history.
• Compare the schedule and actual agents in activity-level (e.g. email, call) views created noise: team leads focused on which agents were out of adherence.

AI opportunities

When this product was designed in 2022, real-time monitoring required continuous human attention. By 2025, AI-driven systems enable a shift toward exception-based oversight:

• Real-time dashboard monitoring may become less critical, because the junior-level monitoring tasks could decrease, allowing team leads to focus on exceptions and complex decisions rather than constant oversight
• Design solution could focus on AI-driven insights can help team leads prioritize actions: suggesting the rescheduling and task reallocation based on real-time data analysis.