Since the emergence of multi-channel self-service, there is a growing disconnect between the typical strategy employed in telephone self-service design and what many customers actually need. This blog details the results of a survey that Nuance commissioned to uncover customer channel preferences for an array of banking tasks, including recommendations for optimizing the IVR to both accommodate those preferences and minimize the frustrations associated with today’s multi-channel servicing climate.

Multi-channel climate overview

Many, if not most, of Nuance’s customers can boast a multi-channel customer contact strategy, ranging from having both web and phone channels to complete suites of mobile applications, email, and SMS strategies. And they should, given the importance of offering multiple channels to both customer choice and convenience.

The problem is, it’s not uncommon for each channel to be viewed as a silo – complete with separate business ownership, different design teams, disparate backend data sources, and disjointed reporting – all of which creates an unsatisfactory experience for customers who transact across multiple channels.

The percentage of customers who willingly use multiple channels to interact with a company is a large majority and growing every year. Simultaneously, customers cite the need to change channels during the resolution of a single task as a key driver for dissatisfaction.1 With the voice channel being the preferred escalation channel, callers are frustrated before even dialing the phone. This puts extra pressure on call centers to successfully adapt to the current multi-channel climate.

Given this pressure, Nuance has increasingly heard demands from customers for a cross-channel perspective on their telephone contact strategies. And if not, it belongs on everyone’s radar. From a broad, multi-channel perspective, Nuance recommends that all businesses approach their call center and/or customer contact roadmaps by:

  • Acknowledging the voice channel as just one of many options, and treating it as an integral part of an overall customer contact strategy
  • Understanding how the customer experience in other channels can have a direct effect on his or her experience within the voice channel

However, despite all the industry publications that point to multi-channel as a hotbed of activity, something has been missing from the conversation. In order to help optimize any contact channel, those responsible require access to more specific detail on the nature of customers’ channel preferences, as well as the causes and motivations for moving from one channel to another (especially when trying to accomplish a single task).

1 Stop Delighting Your Customers – Harvard Business Review, July-August 2010

2 2011 Will Challenge The Status Quo Of eBusiness Online Customer Service – Forrester, January 2011

What channels do customers prefer?

In order for Nuance to continue providing legitimate and actionable recommendations for telephone self-service design which reflect the phone’s role in the greater multi-channel ecosystem, it became clear that ample primary research is this area is needed. Based on our own experiences as customers, a little common sense, as well as on patterns seen across existing multi-channel contact center strategies, Nuance believes that for different customer service tasks customers will gravitate to different channels. However, what’s unknown is which channels do customers actually prefer for which tasks? As a first step, Nuance’s survey partner VocaLabs was invited to help design and host an “attitudinal survey” for the banking/retail credit card industry. In the survey, participants were asked about their preferred channel for various tasks within the domain.

To be clear, the survey didn’t ask participants where they do these tasks today; it asked where they’d like to do them. This is grounded in the assumption that whatever channel customers prefer for a given task, that channel should both address the task and optimize access to it.

The survey results show an interesting disconnect between the typical strategy employed in telephone self-service design and what customers actually need. As you look at the chart below, first notice how routine tasks of checking a balance or getting transaction history -tasks that are often front and center in the typical banking IVR – are perceived by customers as primarily an online activity. Meanwhile, for the tasks that are generally more complex or urgent, such as asking for a credit limit increase (CLI) or disputing a charge, customers gravitate to the channels staffed by humans: the phone or the branch. Yet the traditional banking IVR often buries these latter functions, usually offering them after the simpler tasks above, and even burying them in second level menus.

Also note that while we included many channel options in the survey (website, phone, branch, mobile application, text message, email, and postal mail) the dominant preference is still for the “traditional” channels: website, phone and in-person. And yes, the website can be considered a traditional channel now that the oldest generation is just as likely to use it as most other brackets, as shown in this age-based view. Additionally, it’s apparent from this analysis of that a new wave of channel shift (from the web to mobile applications) is underway – so it’s crucial to continue supporting and refining all channel offerings.

Businesses have a New Set of Challenges to Solve

The preference data from this survey presents businesses with a new set of challenges. A best practice of IVR menu design is to offer the most frequently requested tasks first. And the set of routine tasks outlined above do still drive most phone volume, even though the web is a preferred channel for them. This makes sense, given that many callers asking for a balance over the phone will be reacting to insufficient information in another channel. However, considering that customers actually prefer the phone for more complicated tasks, why should these time sensitive issues not be equally accessible?

The ideal solution is to strike a balance by eliminating the traditional menu hierarchy (via conversational speech) and to improve the handling or routine inquiries to address the root cause of certain escalations (via robust personalization).

Conversational Speech

In order for an IVR to satisfy more customers, the frequency of simple requests should not outweigh the urgency of complex ones. By allowing callers to use conversational language while expressing intent, companies can save their customers from a labyrinth of menus by letting callers describe issues in their own words and routing them straight to the appropriate destination (automated or agent-based). This approach, which uses a statistical language model and is often referred to as “Natural Language,” simultaneously cuts down on the caller’s need for navigation and having to figure out which menu options will best address their needs.

An IVR that understands conversational speech can also discern more tasks and transactions than would otherwise be practical to inventory in a menu. This is important because the increasing use of new channels will continually generate reasons for customers to escalate (e.g., “I don’t have my web password” or “The mobile app on my phone won’t launch”).

Robust Personalization

Enabling personalization is not a simple task. It requires common back ends across multiple touch points, plus a way to store and access profile and history information as customers move between them. The benefits, however, are increasingly compelling as multi-channel use grows. Not only can knowledge of a customer’s prior interactions and account status facilitate quicker routing to the right task, but using it to embellish simple account balance or transaction history information with more detail might provide the “missing link” that drove the caller to escalate. For example, if a caller disagrees with the balance displayed on their smart phone, they may place a phone call in lieu of navigating tiny screens to find the discrepancy. If they are forced to navigate the IVR, as well, that’s hardly a solution. And if the IVR simply reads the same balance, it’s just adding another layer to the frustrating escalation process. But\ if the balance readout contained customized verbiage such as “this balance reflects the $10 fee for your check reorder” or “we also show a pending direct deposit that has not yet posted to your account” the caller’s needs will more likely be met.

What Motivates Consumers to Change Channels?

In addition to learning about basic channel preferences, it’s also important to understand the key drivers for switching from one to another in the midst of a task. When asked to reflect on recent memory, it turns out that almost half of all survey participants who escalated from a preferred to a non-preferred channel did so because of inherent Functional limitations within their initial choice.

The reasons for these limitations are many, but common themes we observed included complaints such as “the task I wanted wasn’t available online” (functional coverage), “I couldn’t get a sufficient explanation” (informational depth), “I had trouble navigating” (channel usability), or “I was told to do it this way” (business rule). Other reasons for changing channels fell in the categories of Channel accessibility (e.g., the customer couldn’t get online, they were stuck at home with an illness, they were traveling far from a branch, or they’d forgotten their password), Channel Availability (e.g., the website was down, the ATM was broken, or nobody answered the phone), and Second Attempt – meaning the customer switched channels after no resolution in hopes of getting a better, or any, response to their problem. While a bank cannot address every reason a customer of theirs might need to escalate from one channel to another, each of these categories also motivates these additional calls to action:

  • Ensure that all desired functions are addressed by whatever channel customers are gravitating to, optimize each channel for ease of use, and keep abreast of new technologies (such as electronic signatures and online check deposits taking the place of in-person appearances at a bank branch).
  • Prioritize resources to keep all channels “up” and available.
  • Many issues related to accessibility are simply out of the bank’s control, but password problems can be addressed by developing effective fallback authentication strategies (including biometric voice-printing) and/or deploying password reset automation.
  • Immediately handle incoming email and “snail” mail, and foster a culture that prioritizes first contact resolution (FCR) rates.

Where do we go from here?

This study is just the beginning of a much-needed look into the way customers want to interact in a multichannel marketplace. There is opportunity to leverage these survey results by comparing a specific bank’s current offerings and usage patterns with a customer-defined ideal. New research is needed, as well, such as to determine the actual amount of channel switching happening at institutions. Nuance expects this will vary based on the quality and quantity of channel offerings available bank by bank.

To address these and other open questions, Nuance is happy to engage with your internal channel stakeholders to craft a long-term strategy for evolving from a silo-based approach to channel management to a consistent, streamlined, and optimized infrastructure.

Let’s work together
Engage us


Let’s work together
Engage us

About Andrea Mocherman

This was a contributed post by Andrea Mocherman. To see more content like this, visit the Customer experience section of our blog.