Increasing scheduling efficiency and transparency

Our customer service agents helped Employee Assistance Program (EAP) clients find and book care. During a live call or chat, they needed to understand the client’s needs, compare suitable professionals, and explain the options—all while keeping the conversation moving.

The legacy booking tool made that work harder. Important provider information was missing, and appointments listed as available could fail when agents tried to book them. Agents had to work around the software while helping clients access care.

A move to a new customer relationship management (CRM) system gave us an opportunity to redesign the provider booking experience. I worked with product, engineering, and operations to make it easier to find a suitable professional, confirm the appointment, and recover when something went wrong. The goal was to support the agent’s conversation with the client from search through booking.

A booking decision depended on more than an open time slot. Agents needed to understand a provider’s credentials, specialisations, and practical details that could affect the appointment. The legacy system lacked important information such as provider alerts and detailed breakdowns of credentials and specialisations.

Availability was also difficult to trust. Agents told us they would often attempt a booking only to discover that the listed appointment was unavailable. They called these slots shadow availabilities.

↗view↗

↗view↗

These problems mattered to operations as well as agents. Our service-level agreements (SLAs) with enterprise clients required us to offer care within short time windows. Helping agents find suitable appointments reliably supported both the client experience and those business commitments.

The first release needed to support mental health bookings. I worked with product to make the case for a foundation that could also support other services, including trauma counselling, financial services, and nutrition support. The immediate booking problem gave us a concrete starting point for that broader opportunity.

Leadership aligned around the larger opportunity, and we secured additional scope and budget to build a foundation to support other services over time.

Operations had identified the need for a better tool. To understand what that meant in practice, I interviewed and shadowed eight customer service agents as they worked.

Shadowing was especially valuable. Booking was only one part of a live client conversation, and agents were often listening, searching, comparing, and explaining at the same time. A flow that appeared simple in isolation became demanding in practice. Agents also had to keep clients informed and avoid prolonged silences, even when their tools failed or they lacked the information they needed.

↗view↗

I documented the workflow and the workarounds agents had normalised, including how they dealt with unreliable availability and moved between communication tools. This helped us distinguish problems the booking experience could address from wider process issues.

We also reviewed internal and external scheduling products to identify familiar patterns. Agents already had working habits built around the old tool; preserving useful patterns would help them move to the new experience without relearning every part of the job.

I mapped a high-level service delivery flow based on our research and goals. We also needed to account for interruptions and errors, helping agents return to the booking workflow with as little friction as possible. The simplified flow is shown below.

The research gave us four priorities for evaluating the design. Agents needed to find a suitable professional quickly, have enough information to recommend them confidently, and complete a reliable booking. We also needed a foundation that other service teams could build on.

Agents needed to compare options while listening to the client, so we kept available appointments in a results grid with filters on the right. Provider details opened in a side panel, allowing agents to investigate a possible match without losing the results. This reduced navigation and repeat data queries while keeping the layout familiar.

Testing helped us simplify that layout. An early concept separated frequently used filters into their own section, but agents found the distinction confusing and sometimes missed the controls. We brought the filters together in one clearly labelled panel.

The walkthrough below shows how agents could narrow the search, review a provider, and choose an appointment while keeping the client’s options in view.

Finding a match was only part of the task. Before booking, agents needed to read the appointment details back to the client and pass relevant context to the counsellor. The confirmation view brought those tasks together. If the selected time became unavailable, agents could review alternatives and adjust the booking within the same flow.

We compared booking errors and rematch requests in the six months before launch with the six months after launch. Both percentages represent relative reductions.

The efficiency gains came from improvements to the interface, underlying technology, and reliability. Reducing average handle time (AHT) was a team effort.

The trauma team also began adapting our technical and UI foundations to its own needs. Adoption was still underway when I left the team, so I could not assess the full rollout. That work showed the potential to extend the booking experience to other TELUS Health services.

The results reflected the problem we had set out to address: helping agents complete bookings more efficiently and reliably. The trauma team’s work also gave us an early indication that the foundation could support needs beyond the first release.

Shadowing agents changed how I evaluated the design. A booking flow had to work alongside listening, explaining, and supporting a client. Keeping information visible and making errors recoverable mattered because they helped agents continue that conversation.

Making those everyday problems clear also helped us build support within operations. We could connect the design priorities to service commitments and show why the work deserved more ambition than a like-for-like replacement. Those partnerships helped us secure support for a foundation other teams could build on.

As with most things, this is a brief look at the project. Feel free to reach out if you'd like to discuss anything in more detail.

I mentioned that we worked with our trauma team on a location-centric booking tool. Below, you can see some of our initial concepts. The prototypes were well received and were designed to address a major challenge for agents: quickly finding a nearby counsellor available for scheduled work while also accounting for emergency situations.

We only really scratched the surface, as there were still more questions to answer around multiple referrals and prioritisation, but I think where we landed is interesting enough to share.