IVR Consumer Research was one of my favorite classes in business school. Having always been a techie, I was amazed to see how science drives seemingly mundane behavior and preferences. I was surprised to read about controlled experiments and test cases being used in the real world to get a better feel for consumer attitudes, behaviors and perceptions. In most cases, this experimentation data was used to improve predictability in business, predict traffic flows, simulate end-to-end service and identify break points—all to ultimately drive the next service improvement. No wonder leveraging consumer focus groups and experimentation is an already well-established practice in the consumer goods industry and is picking up considerable momentum in the service sector!

Using Consumer Research to Build Better IVRs

If you are an enterprise client planning to invest significantly to create fulfilling consumer experiences through an IVR, to augment existing service options or to completely remodel your IVR, wouldn’t it be worthwhile to get a sneak peek into your consumer reaction? Post-launch optimization is a good practice but if you are a business handling tens of thousands of calls each day, wouldn’t pre-launch improvements give you a significant leg up?

Instead of going all-in with investments and risking customer backlash, consider instead implementing a scaled-down version of the actual site and exposing it to different consumer focus groups to study their preferences. You could experiment with different IVR call flows, voice talents, scripts framing, sequence of options, etc., and use that feedback to drive the most effective IVR.

How to Experiment with Your IVR

Depending on your target demographics and characteristics, you need to make sure that your focus group includes a representative sample of your target audience. You can further divide them into groups and emulate scenarios—artificial conditions—to see how they interact with the system under different situations. One of the many advantages of such experimentation is that it may be easier to establish causality as the experimenter can vary one feature at a time in a controlled environment. We can vary just a few IVR parameters such as menu sequence, sentence framing and wait between responses, then analyze what is the most effective design. The definition of “effective” would be very dependent on your business, what the IVR is intended to do and how consumers perceive that experience.

You can emulate traffic flows, find dead zones where users hit roadblocks and predict call center transfer traffic loading and wait times based on experimentation data. Having more choices definitely helps, and with more IVR design options and performance data to back it, businesses should have a fair amount of confidence in consumer acceptance. Better predictability will create better resource planning and improve complete service line.

At Angel, we believe that experimenting and creating “beta” voice sites should be streamlined and easy. This means being flexible for changes on the fly and being ready to deliver updated versions to focus groups as soon as they’re ready. It also means we have put a lot of time and effort into our CX Analytics, which works hand in hand with the SiteBuilder tool to give our clients that much-needed insight into popular paths and dead zones so they can optimize call flows accordingly.

Learn more how to create a top performing voice applications by downloading, Driving Performance With Embedded Business Intelligence Analytics.