Reimagining the Contact Center Experience
In 10 weeks of Collaboration between Deloitte’s Government & Public Services and SCADpro team made up of 14 students, we were challenged to research and develop concepts for government contact centers in the US to overcome user bias that exists today between interactive voice response technology and callers.
Role
UX Designer, UX Researcher
Years
Sept 2021 - Dec 2021
Company
Deloitte and SCADpro
Location
Savannah, GA
Project Breakdown & Approach
Secondary Research - What is the current state of the IVR system?
IVR as we know it today first appeared in the 1970s and its popularity among businesses and government organizations soared in the 1980s; however, despite being around for decades and achieving near universal popularity among organizations, IVR systems today remain largely unused by callers. However, the technology is not only facing a lack of adoption but is also competing with an accumulation of negative biases from its extended history and the cutting-edge capabilities of novel technologies like smart assistants.
Approach to secondary researchWhen tackling secondary research as a team of 14 students, it's vital to establish clear communication and designate specific roles. Our team individually gathered data points found from case studies, articles, infographics, videos, and more. This research led us to 530 data points and 71 sources. As a leader on this team, I encouraged open discussion and conducted regular progress check-ins during our focused time of gathering secondary research.
Primary Research - Why don’t people like Contact Centers?
One of the main questions ideas we wanted to understand better was the distaste for contact centers. We learned that top pain points included: The reason for calling might not be listed (65%), being forced to listen to irrelevant options (63%), IVR keeping callers from reaching a live person (54%), and the menus are usually too long (46%). 52% of people were frustrated at having to repeat themselves, while 49% hated waiting to be connected to a live agent while in a queue.
Approach to primary researchOur team called 15 businesses that currently use IVRs to analyze the different patterns that matched the points we found during secondary research. From here we specified our questions and I led the creation of a 17-question survey. From the 93 survey responses, we gathered 486 data points. Our team then conducted 16 interviews with people who have contacted help centers to gather an understanding of what could help remedy this frustrating experience.
Synthesis and Research Findings
The main demographic within the group of non-adopters is older generations who have difficulty acclimating to the rapid changes in technology. With primary and secondary research we found out that this group prefers human connection and misses the empathy a ‘machine’ can’t provide. This target group finds the IVR menu options overwhelming and generally starts IVR interactions with a negative bias and doesn’t think the IVR can solve their problem.
Approach to SynthesisWe affinitized our data by organizing common themes from data points in primary and secondary research, we then wrote detailed summaries of each theme and categorized the themes. After open discussions about the categories we organized, we felt confident to write out our primary research insights. We understood user pain points, biases against IVRs, adoption issues, and ways people are looking to have more human experiences.
Defining user biases around IVRs
Biases lead to adoption issues which cause users to avoid the IVR and want to go straight to an agent. Factors that cause adoption issues are the IVR not being personal enough, the IVR system being inefficient, and not being flexible enough for users’ needs. We defined 7 leading biases users have against IVRs. They are: (1) Some people don’t think they’ll be able to use the technology effectively, (2) IVR is slower than an agent, (3) IVR will not be able to solve my problem, (4) IVR interactions are impersonal, (5) I need human connection, (6) IVR voice and dialogue feels and sounds untrustworthy, (7) IVR is difficult to understand and communicate with.
Approach to Defining biasesWe pulled data from our findings that were focused on biases and their contributing factors and grouped the behaviors we observed in those biases. We discovered in our research that users would like IVR to be more concise, flexible, and accurate in its responses, as well as have personalization that can aid in making the service more concise and helpful. We understood the reasons why people have these biases and were confident to state the main biases against IVR adoption.
Solution and Iteration
For our final deliverable, we created 10 total recommendations, highlighting the top 5 that had the highest confidence.
Approach to SolutionTo validate our recommendations we conducted user testing. Our testing started with an audio-based survey to narrow down and refine our most effective recommendations. 21 questions were asked with 143 completed responses. In order to assess the relationship between people’s preference for our recommendations and the adoption of technology-delivered solutions, we also performed a series of one-on-one user tests. We designed three prototypes to determine the relationship between preference and adoption: a best-case scenario consisting of the five most preferred recommendations, a comparison case consisting of the 5 least preferred recommendations, and a control case representative of current IVR designs. Voice-related recommendations were excluded from the recommendations used in the prototyping due to the significant impact voice has on the experience, creating a need to test these recommendations independently.
Results
Our team’s recommendations and research were featured in an article written by Deloitte. Find the article below!
Happy to chat more about my role as a designer and researcher on SCADpro and Deloitte collaboration.
Linkedin
Alejandra Rodriguez Profile
Phone
(210) 255-9102
Email
jandrarod2@gmail.com