Member IVR Call Flow Abandonment

Project Overview

The vast majority of members who call into Centene do not try, or are unable, to “self-service” in the Interactive Voice Response system. Moderated user interviews were conducted to gain deeper insight into participants’ responses and lack of engagement with the IVR flow. Due to constraints with UserZoom’s technology (the inability to host verbal recordings on its platform), voice files mimicking the IVR flow were used. ​

Objectives

  • Gain insights from members to understand the behavior behind the numbers.
  • Identify possible weaknesses of our IVR
  • Identify opportunities for improvement

Only 4.92% of members successfully self-service in the IVR

Tools

  • UserZoom
  • Excel
  • Miro
  • Microsoft Powerpoint

High-Level Research Process

The IVR team was new to having a dedicated UXR and more accustomed to conducting evaluative research, so this was a great opportunity to show the impact of generative research. I encouraged them to think about how this exploratory research could help us shape the design of our IVR system.

  • I met with my key PM/design stakeholders for a kickoff call to gather information on the study which helped me fill out my research study plan template with specifics on this study (including statistics from PowerBI, key prioirities for the IVR system, key stakeholders, summary, background, related research, business goals, hypotheses, research questions, methodology, timeline, recruitment/participant requirements, communication preferences, deliverable, next steps, instruments, out-of-scope/future studies and kickoff notes).
  • We determined key participant demographic information to collect (age range, type of insurance, methods used to manage health insurance, such as mobile app, website, or phone, and their role in coverage decisions). Then, we recruited a total of 12 participants with an even split between the primary decision makers and those who share equally in the health decision process.
  • I shared the finalized study plan and screener with my team so they could provide feedback, with deadlines for feedback and the expected date of the final deliverable.
  • Then, after sign-off on the study plan and screener, I started on the moderation guide for this study and sent it to them for feedback.

High-Level Questions

  • Can the user easily navigate/understand the call flow?
  • Why are our callers failing to self-service? User preference? Is the call flow difficult to understand?
  • Are we missing vital information?
  • Is there too much information?
  • Is the pacing easy to follow?

Data Collection & Synthesis

  • The study methodology consisted of qualitative interviews at 45 minutes long, where I would play voice files.
  • Data collection lasted two weeks, with rescheduling and cancellations.
  • Synthesis was ongoing during data collection, with design stakeholders included in the process.

Deliverable Deployment & Readout

The final deliverable was decided upon during study planning with the team, which we agreed would be a deck with key quotes and recommendations for prioritization of specific features to improve the call flow to retain members in the IVR system.

Impact

This generative, exploratory study shaped the revamping of the call flow process, including changes in the location and language used in the presented options.