Getting Customers Out of the Queue
How a financial services provider used a design sprint to reinvent its IVR system in just five days
Let’s be honest. We are all pressed for time these days. No matter what we do, where we are or what channel we use to communicate, as soon as we encounter a problem, we want it to be solved immediately. This is especially true for customer service. Customers don’t want long waiting times, they want immediate and straightforward answers.
But what happens if an organisation seems to have the right technical set-up, but the service still doesn’t work smoothly? Where do you start? How can the customer service processes be optimised in the shortest possible time?
This was precisely the challenge faced by one of our clients, an international insurance company. Despite routing calls by means of a modern interactive voice response system (IVR for short), customers were still frequently forwarded to the wrong service agent. This resulted in numerous redirected and duplicated calls while also impairing the accessibility of the service agents – issues that were not only frustrating for the staff, but also greatly impaired the customer experience.
A quick yet comprehensive solution was needed, making this the ideal use case for a design sprint approach.
Day 1: Understanding the problem
Around 40% of enquiries to the company’s customer service don’t end up with the right agent and have to be forwarded.
After a short introduction and getting to know each other, we started by conducting expert interviews that had already been prepared in advance. These interviews were conducted with the following internal and external experts:
- a customer service team leader,
- an internal service manager,
- an IVR project manager from the claims department,
- a process manager,
- a technical IVR manager and
- an external IT expert from the company providing the customer service software.
We then grouped the issues, insights and ideas that emerged from these interviews by theme. Together with the knowledge of the design sprint participants, this served as the basis for developing a map depicting the various stakeholders and the existing process. Finally, we also defined the long-term goal of the design sprint:
«We want to make sure that customer calls are routed to the most competent agent».
At the end of day 1, the problem could be summarised as follows:
About 40% of telephone enquiries to the company’s customer service had to be forwarded to a different agent. Many of these enquiries fell into the category of «other issues». Why? On the one hand, some important categories were missing from the IVR. And on the other hand, customers were often unable to (correctly) identify their request. In addition, there were also customers who somehow «clicked» their way through the IVR, because they were unwilling to wait in a certain category or wanted to be quickly connected to a service agent.
Day 2: Gathering ideas
Adopting an outside perspective, discussing the issue within the team and coming up with own creative ideas.
After we had understood what the main issues were and decided what to focus on in the route map, we looked at alternative solutions:
- How do other companies cope with these problems?
- What ideas already exist in the market?
- Which ones have proven to be particularly successful?
After an in-depth exchange of ideas that incorporated the findings from the expert interviews, we then turned the tables. Now each of us was tasked with developing our own creative approach to solving the problem.
Day 3: Determining the direction
Which ideas harbour the greatest potential?
On the third day, it was time to make the first decision:
Which ideas did we want to pursue further and which ones not? What kind of approaches would meet both the needs of customers and the interests of the company?
The decisive factor was whether these ideas could also be implemented.
After the presentation of the ideas and a vote by means of a heat map, the following approaches came out on top:
- Usability for customers: Improve the clarity of the IVR by using the shortest possible text in simple language. And adding additional categories to the IVR that were previously missing.
- Making queries more flexible: Quick, ad-hoc additions for queries that occasionally arrive in large numbers, such as «hail damage», enabling better routing of customer queries.
- Data enrichment: Linking the IVR with the CRM data so that language settings can be anticipated, for example, which makes it possible to shorten the IVR announcement.
- Alternatives to IVR: Provide alternative options so that customers do not have to listen to the entire IVR. In this regard, we pursued two ‘dark horse’ ideas . On the one hand, listing direct phone numbers on the website in line with the content of the specific page (e.g. car insurance – the direct sales phone number for car insurance) and, on the other hand, providing customers with a clickable version of the IVR.
- Flexibilization of queries: Quick ad hoc additions for requests that are temporarily frequent, such as «hail damage». In this way, customer queries can be better routed.
- Data enrichment: linking the IVR with CRM data so that language settings can be anticipated, for example. This allows the IVR announcement to be shortened.
IVR alternative: Provide alternative options so that customers don't have to listen through the IVR. In this regard, we pursued two 'dark horse' ideas. On the one hand, direct phone numbers should be offered on the website according to the website content (e.g. car insurance; phone number from direct sales for car insurance) and on the other hand, customers should be provided with a clickable version of the voice control.
Based on the four leading ideas, we then created a storyboard in preparation for the prototype development, which took place on day 4.
Tag 4: Developing the prototype
Things are getting serious.
After allocating the tasks and setting the deadlines, our first order of business was to work on a prototype for the new IVR structure. The storyboard already contained three different approaches, which now had to be elaborated in separate teams. The voice texts differed in terms of structure, word choice, content hierarchy, topic categorisation and length.
In the afternoon, we divided the team into two groups. While one group put together the IVR texts using a computer- generated voice and then edited them accordingly, the other created a clickable mock-up of the website and used it to implement our IVR alternative. The IVR prototype ultimately consisted of several audio files that made it possible to experience the new IVR structures in an audible way. Meanwhile, we created a mock-up of the website using print screens, which we had animated and modified by means of a Power- Point presentation. The aim was to make sure that during the customer testing on day 5, nobody would notice that our solutions existed only as a prototype.
Tag 5: Customer testing
«Payday» – honest, direct and immediate customer feedback
Through an external agency, we had recruited five test customers who were now called upon to put our prototype through its paces. The test took place in several stages. An interviewer presented the prototypes to one of the test participants, while the rest of the team followed the interview in an adjoining room and took notes on the key takeaways.
These test interviews were an educational experience for the whole team. The new IVR logic performed significantly better than the existing system. Nevertheless, the test participants were not really enthusiastic about the IVR. «I don’t like to wait» was the general consensus among them. In addition, they also emphasised the importance of human interaction.
At the very end of the test interviews, we presented the website mock-up to the test participants.
This was a great success, the testers were highly enthusiastic. «This is exactly how I’d imagined it», as one of them put it, while another called it «excellent – very customer-friendly». The direct phone numbers on the website and the clickable IVR on the contacts page were a hit with the participants. All test participants thought that the website was real, not least thanks to how well it was presented.
The design sprint was a success and had a direct, positive impact on the insurance company. We were not only able to develop a more customer-friendly IVR structure but also offered customers an alternative to the phone-based IVR available on the website. At the same time, the effort and costs associated with implementing the new solution was manageable. The new IVR and the direct contact options on the website led to a measurable reduction in the number of referrals. Customers are much more satisfied with the new solution and it also saved the insurance company money.
As we found out later, there was an additional unforeseen but welcome side effect: Listing the direct phone numbers on the company website meant that they showed up in Google searches. This significantly improved the ranking of the website in search engine queries.
- Are you facing a similar challenge?
- Are you looking for ways to increase sales or attract new users?
- Would you like to improve your processes and save costs?
- Do you want to improve your product or service?
- Do you have a business ideat that you want to market quickly?
- Are you looking to start a new project and would like to identify the right strategy?