Too many IVRs are developed without giving due thought to the callers who use it. Here’s why it pays to do user experience (UX) research.
Steve Jobs famously said:
“You’ve got to start with the customer experience and work back towards the technology - not the other way around.”
Yet for many companies designing and developing interactive voice response (IVR) systems, that user experience is an afterthought - if it’s considered at all.
The result is IVRs that do all the things disgruntled customers complain about. They present confusing menu options, they use impenetrable jargon, they sound inhuman and robotic, and they cut people off mid-call.
In short, they’re the very opposite of a rewarding brand experience.
So how can you avoid creating an IVR that plunges your customers into phone menu hell? That’s where user experience (UX) research comes in.
The risks of not doing IVR UX research
At VoxGen we’re often asked to review IVR applications that are failing to deliver against customer service targets. Very often, we find that those systems are failing because little or no research has been done into what the users of the system - the callers - actually need.
Without that insight, it’s easy to design a system that reflects business goals, but doesn’t give users the kind of experience they want.
A good example is an IVR that puts the “place an order” menu option first, because the business wants to make sales - ignoring the fact that most calls are from customers checking when an item they’ve ordered will be delivered. So callers have to wait for that menu option to come up - and the longer they wait, and the more irrelevant options they hear, the more frustrated they get.
There are tangible downsides to this: at one insurance provider that we worked with, the IVR had basically become unusable, causing high levels of inbound calls and costly internal transfers. And there are the less-tangible downsides, too, like customers posting angry tirades on social media.
Five best practices to get the most from your IVR UX research
So it pays to pay attention to your users, and set time and budget aside for professional UX research. There are lots of best practices you can adopt to get the best results from that research. Here are five that we use on every IVR project we work on:
Integrate IVR UX research throughout the design and development process.
UX research isn’t a once-and-done exercise. Yes, you need to do early context gathering to identify user needs, but you also need to get user feedback throughout the design and development process, so you can be sure it’s always heading in the right direction. Getting it right early will reduce the amount and cost of any rework later in the project cycle.
At VoxGen we use an iterative approach consisting of three main activities:
- Context gathering to identify user requirements. We do this using a range of techniques including call listening, usability evaluation of the current IVR, interviews with stakeholders, agents and users, and analysing IVR performance data
- Iterative design and usability evaluation. Based on the context gathering, we’ll build a prototype system, test it with representative users, then refine the prototype and test again until we have a usable model. This ensures that usability issues are understood and addressed early in the process - before development begins.
- Ongoing research. Once the system is live, we will conduct “in live” UX research to ensure continuous optimization and governance. That research can be done using surveys, performance analytics, user interviews.
Understand IVR user requirements in the context of the end-to-end customer journey.
Lots of companies think of the IVR as a standalone application. But in reality, callers move between different channels and devices, and IVR may only be part of their overall journey.
By understanding how and why callers move between (for example) the website, the mobile app and the IVR, you can understand where IVR fits in the overall context of how they engage with your business. This will typically involve a formal process of customer journey mapping - which should take into account all of the context gathering done with real users, rather than making assumptions based purely (for example) on system analytics.
Customer journey mapping will help you identify which tasks callers are happy to undertake in the IVR - and which tasks might be better suited to another channel. It’ll also give you vital insight to help you optimize cross-channel experiences.
Involve everyone in the IVR UX research.
IVR is a key customer touchpoint, and one in which lots of people tend to want a say. Marketing may want to include marketing messages, Sales may want to prioritize “place an order” type options, and contact center managers may want to stave off inbound calls by - for example - not including an option to speak to an agent.
That can lead to conflicting demands and a lot of pressure on the design and development team. But when everyone is involved in the research process and can see user feedback first hand, they’re able to focus much more clearly, as a team, on what the user needs from the system. That makes for a more harmonious project, as well as a much better caller experience.
The key here is to appoint a UX research professional to plan the research, communicate and interpret it for stakeholders, and co-ordinate the response to it. It’s essential to establish the right research approach and governance framework for your organization, as the wrong approach can lead to design decisions that damage, rather than improve, the user experience.
Don’t forget about IVR UX after deployment
Even when UX research has been done at the start, sometimes all thoughts of the user go out of the window after the IVR goes live. (This isn’t helped by the fact that many conventional IVRs provide very little in the way of analytics.)
That leads to a siloed, bolt-on approach to IVR changes, with marketing teams, sales teams and call center staff inserting messages and menu options to meet immediate business needs, without considering the impact on the overall caller experience. (Who hasn’t called a customer service line only to hear a hastily-recorded and now out of date message about an outage last month?)
If you want your IVR to remain useful and helpful for your callers over the long term, you have to continuously optimize it in light of continuously-gathered user feedback. That won’t happen on its own: you’ll need a governance process to ensure a structured and informed approach to IVR optimization, with all stakeholders involved and informed by UX research.
Measure the IVR end-user experience.
At the same time, you don’t want to be conducting focus groups every week. For ongoing UX research, you’ll want to use relevant metrics and set easily-measurable KPIs that are specific to IVR interaction.
There are lots of great metrics around to help you measure the quality of the customer experience, and you may already use a variety of them across your organization. From an IVR perspective, though, some provide more meaningful insight than others. NPS, for example, is often used to measure a customer’s propensity to recommend the business to others, but that kind of consideration is unlikely to be driven by the IVR alone. We find that CSAT is a much more meaningful metric for IVR, as it measures a customer’s satisfaction with the specific experience they just had.
However, the main thing is to determine what it is you want to measure, determine the KPIs you’ll use to measure it, and set a baseline to measure improvement from. Then it’s a case of measuring the performance, understanding what’s driving any changes (by conducting further qualitative UX research), and optimizing the IVR accordingly.
IVR UX Research: Cultural Considerations
One of our current research projects is investigating the impact of Cultural Differences on design....
What We've Learnt From Years Of IVR UX Research
The UX (User Experience) industry has seen some big changes over the years. I began my career in UX...
Is your IVR menu giving your callers indigestion?
Your IVR menu may make perfect sense to you - but is it confusing your callers? If so, it’s time to...