Can You Hear Me Now? Top Five Voice of Customer Pitfalls


Share on LinkedIn

An Open Letter to My Suppliers

Hi there! I’m one of your customers. Sorry to interrupt, I know you’re really busy, what with the down economy and all. Must be tough cutting staff while at the same time trying to get your customers to open their wallets.

Anyway, let me get to the point of this letter: your Voice of Customer (VoC) programs. Over the past couple of years it seems most of you have jumped on the VoC bandwagon. Every time I turn around, I’m getting another survey invitation by email, phone or even the mail.

I think it’s great that everyone wants my feedback—listening is a good start towards building a real relationship. But honestly, who has the time to fill out all those surveys? It seems like you’re missing some new options to really hear what I’ve been trying to tell you, without asking me to rate everything on a scale of 1 to 10.

Instead of filling out all your surveys, I decided to write this letter about VoC pitfalls, based on what I learned from companies that seem a bit more experienced. I’m crossing my fingers that you’re really interested in what I have to say, and might even make some changes as a result.

Pitfall No. 1: Lack of executive support to drive change

Many years ago, I used to be a Symantec customer. I really liked the Norton antivirus product—simple, effective and priced right. But then Symantec started making more complicated security bundles that were hard to install and use. The software did such a good job protecting me it just ground my PC to a halt. So I switched.

But recently I bought a new laptop for my son Matthew, after he learned that dousing the old one with milk doesn’t make it work any better. The new computer came with Symantec pre-installed, so I decided to give it a try. And what do you know, it was a much improved product.

Apparently I wasn’t the only one that had issues with Symantec software. Desirree Madison-Biggs, the firm’s Director of Customer Experience, told me that despite “customer-driven” being a core value, many customers felt that Symantec was more technology-focused. So, four years ago she was charged to implement a VoC program for this huge company with 55 million paying consumers and tens of thousands of enterprise customers.

With the help of Satmetrix, an Enterprise Feedback Management (EFM) vendor, they developed a method to systematically find and communicate customer issues in a common language that everyone could understand. Especially the CEO. A Net Promoter Score (NPS), based on the “Would you recommend us?” question, was used as a kind of “north star to reflect how we’re doing from a customer point of view,” says Madison-Biggs. They also asked other questions and analyzed comments to figure out what needed fixing.

The impact of the VoC program has been impressive, because Symantec actually did something with the feedback! With the recent Norton products, installation failure rates decreased from 11 percent to .3 percent; installation time reduced 500 percent to just 43 seconds; and performance improved substantially.

Still, when I asked Madison-Biggs to sum up four years of hard work with her best words of advice, she didn’t talk about scoring methodologies or technology. Instead, her advice is to make sure that top executives “believe with heart and soul in the importance of a VoC program and then drive real cultural change.” At Symantec, executive support starts at the very top with John Thompson, the CEO.

Pitfall No. 2: Garbage in, garbage out

Computer geeks know that if you put bad data into a perfectly designed computer program, you get perfectly useless results. That’s known as “garbage in, garbage out.” Of course, you also get garbage out if the data is good but the program has logic designed to tell you “1 + 1 = 3.”

Instead of blindly using a metric you read about in a book, work with firms that can help you not only collect feedback data, but also build a valid VoC model.

With VoC programs, you’re collecting customer feedback (input) so you can get insights you can act on (output). If you ask customers the wrong questions, that’s just garbage in. If your VoC program is built on faulty logic about what really impacts customer loyalty, you’ll waste time and money making changes that don’t matter, or actually make things worse. (For more on this topic, read my article Find the “Ultimate” Loyalty Metric to Grow Your Business.)

Consider Chilis, a casual dining restaurant chain owned by Brinker International. Employees serve more than one million guests every day in four restaurant brands with more that 1,700 restaurants around the world.

Mark Williamson, Senior Director of Guest Experience and Hospitality, has the charter to measure the guest experience and provide direction on how to “create passion” for each brand. What that boils down to, he says, is a combination of emotional engagement plus the dining experience. At Chilis, Williamson found that while good food is essential, the key differentiator was in fact the “hospitality” of team members.

With the help of Empathica, an EFM provider that provide both technology and research services, they found a “very clear linkage” between employee behavior and the guest experience, which was in turn was related to increased restaurant traffic. So, Chilis’ management works hard to hire people with that all-important “hospitality gene” and ensure they deliver a great dining experience.

Instead of blindly using a metric you read about in a book, work with firms that can help you not only collect feedback data, but also build a valid VoC model. As Williamson says, “don’t be so arrogant as to assume you know what customers want.”

Pitfall No. 3: Employees aren’t motivated to be customer-focused

Employees are people and tend to do things in their own self interests. Surely you know that, it just seems so obvious.

Why is it then, despite all of your VoC activities, that some of your service reps rush me to get me off the phone when I call for help? One time, a call center agent apologized, explaining that her bonus was tied to having the lowest “average handle time.”

Are you really saving money with these shorter calls? Let me tell you a little secret—after I hang up on a call that didn’t solve my problem, I just call back or use some other support channel. Or worse, I take my business elsewhere and tell my friends about my lousy experience. Stick that in your ROI spreadsheet!

If you were really listening to me and other customers, you’d know that in most cases we just want to get our problem solved with one call. So why don’t you measure employees on “first contact resolution” (FCR) and reward them accordingly?

In other words, take a tip from Zappos, a popular retailer founded 10 years ago as an online shoe shore. Since then it has expanded to sell handbags, clothing and electronics to a very loyal customer base. What’s different about Zappos is the attitude of employees towards customers. In the Zappos call center, “customer loyalty representatives” are measured on FCR and rewarded for the quality of conversations, not speed.

Another great example is Mountain America Credit Union (MACU), the second largest credit union in Utah with over 300,000 members. According to a recent Bancography study, MACU is the strongest large credit union brand in the country. Why? Because management works hard to create a positive work environment for employees, and engaged employees lead to engaged customers, according to Senior VP Brent Lawrence.

Research revealed that the key factors in employee engagements were worker’s pride, relationships and recognition, work/life balance and a supportive culture. Using employee and customer feedback collected with Allegiance, the credit union monitors trends in engagement scores and takes action when needed. In one case, they resolved a member attrition problem at a branch office by fixing the root cause: declining employee morale.

If your employees are happy and motivated to do the right things for me, there’s a better chance I’ll do the right things for you. Like spend more money.

Pitfall No. 4: Listening with only one ear

While it’s easy to analyze quantitative feedback generated by asking me to score things on a numerical scale, that’s like listening to customers with only “ear.” The other ear should be used to understand customers through the unstructured feedback they provide.

Use text analytics to listen to customers on their terms, so you can act on emerging issues long before you could field a conventional survey.

Comments on surveys are a natural place to start, but customers also send in text feedback via web site forms, email messages and chat messages. Call center agent logs, transcripts from recorded phone calls and posts on blogs or discussion forums are also good sources for customer feedback, if you’ll take the time to read.

Yes, reading comments is time consuming. But hey, maybe if you read more customer feedback, you’d have a better idea of which problems to fix, which would free up some call center staff time. Just a thought.

JetBlue gets 45,000 comments per month, says Bryan Jeppsen, the airline’s customer feedback analyst. Until recently, they made business decisions based on “reading as many comments as possible”—about 300 per day per analyst.

Automation was introduced as a result of the infamous New York ice storm of 2007, when JetBlue was overwhelmed with 15,000 emails in just two days. Attensity, a text mining vendor, stepped in to help analyze the messages. Customers were upset about the delays and cancellations, and disappointed that JetBlue didn’t have a backup plan. On a more positive note, however, some customers wrote to compliment airport staff and in-flight crews on their handling of a difficult situation.

Since that crisis, JetBlue has worked on a pilot project with Attensity which Jeppsen says helps mine customer sentiment, as well as providing “tangible data around how to augment JetBlue services.” By tying feedback data to a specific aircraft or even a seat number, they can find and fix problems that have a direct impact on the customer experience.

If your customers are suffering from survey fatigue like me, shorten the surveys and provide more opportunities to provide comments. Expand other options to provide comments, via your web site, chat sessions, etc. Then use text analytics to listen to customers on their terms, so you can act on emerging issues long before you could field a conventional survey.

Pitfall No. 5: Ignoring social voices

As you may have heard, consumer usage of social media has exploded in recent years, including blogs, review sites, Facebook and more recently, Twitter. Lots of options to rave about great experiences or vent about bad ones.

VoC Solutions

Now, I fully appreciate that social media is a chaotic and noisy world. Hard to “separate the wheat from the chaff,” as they say. And besides, how do you know if the complainers are really your customers?

Despite these challenges, can you really afford to turn a deaf ear to social voices? JetBlue is already using Twitter to listen to customers and respond to service issues. In the future, the company plans to use Attensity’s MarketVoice solution to more systematically mine all sorts of consumer-generated content.

Stephan Sigaud, a noted expert in the field of loyalty research, agrees that “social media is an emerging feedback channel that companies cannot afford to ignore.” I think it’s the wave of the future, because, frankly, it’s a lot easier for me to post something on Facebook or Twitter. Have you tried to navigate your company’s web site lately?

Well, that about does it for my feedback on your Voice of Customer programs. Avoid these pitfalls and you’ll really hear what your customers are saying, including me.

Then you’ll know what you need to do to encourage us to spend our scarce funds with your firms in the future. Don’t get discouraged, the bad economy won’t last forever.

Thanks for “listening,” and good luck!


  1. This was an excellent piece Bob and I enjoyed reading it. I think far too many companies are “ticking the box” of customer intimacy and, in reality, doing very little. At best they’re collecting feedback but either not acting on it or not closing the loop with the customer.
    One point on Attensity (and all the other mainstream text analytics vendors) – they all start with a categorisation structure (that is time consuming and costly to create) into which they categorise the unstructured data. This means they are determining the outcome before the analysis – which introduces bias and prohibits the discovery of unkown unknowns.
    The technological answer to automate all the disparate text analytics requirements in this space (including social media etc) doesn’t exist yet…I’d like to think that we at Top Right Corner can launch that solution in 2010.
    Regards, Neil

  2. Your article, as always, is right on target. Companies can avoid many of the pitfalls you identify – actually not even experience them in the first place – with a little research-based discipline. I’d like to address two key subjects covered in your article: impact of employees and social media.

    Employees have a great deal to do with optimizing customer experiences and relationships. We feel that an easy, and revealing, method of identifying degree of synch with customers is through ‘mirroring’. This technique has employee rate and rank elements of performance as they think customers would; and then their results are compared to almost identical research with customers to see where there is agreement or disagreement. We also have developed an exclusive framework for understanding employee commitment to the customer, to the organization, and to the product/service value proposition.

    Social media, and analyzing sources of informal communication as well as their impact on downstream customer behavior, is becoming increasingly important. We’ve actually developed six different methods for helping understand the effect of communication through customer-generated media – everything from text mining/analytics and web content harvesting to identifying where and why information is gathered and how it influences customer selections. In many industries studied through our ‘Customer WOM Power’ research technique, positive communication is both more frequent, and more powerful, than recommendation at leveraging behavior.

    Michael Lowenstein, Ph.D., CMC
    Senior Vice President and Senior Consultant
    Harris Interactive

  3. Too often, we as the customer want they want/need. They only feedback what they know. It is crucial to understand behavior and being able to translate behavior into action (we are the product/service experts).

    It like asking consumer what they needed before the advent of the minivan. They would have said a bigger station wagon. However, behavior drove the need for creation of the minivan.


  4. Bob: Your article is spot on. And as always, I read it with great interest.

    I cannot agree more on the piece you’ve written about employees and the role they play in customer centricity. Linked to that is a point that many companies miss out, when they create customer satisfaction questionnaires. They ask or try to find out about customer experience related to a recent experience and work on a satisfaction scale based on that. The employee and the interaction they’ve had with the employee is not taken into consideration at all.

    And if they did, its too superficial to even arrive at any meaningful conclusion. I think its important to seek feedback on

    1. Employees motivation levels to get things right the first time
    2. Employees motivation levels to go the extra mile to serve the customer
    3. Levels of employee empowerment to resolve issues without much escalation
    4. Consistency of employee behavior / engagement within and across various touch points
    5. Alignment of employee behavior with organization’s customer service / customer centricity goals.

  5. A great article Bob – very readable. My experience is that where companies are moving into gathering Voice of Customer feedback, they have typically come from a traditional market research perspective, and so tend to organise the feedback gathering in a very structured way. This can cause some of the problems you mention about the way that the information can then be analysed and acted on. One of the main areas that VoC feedback can help is to substantially cut down on the length of a survey or questionnaire, where one simple open-ended question can cut out dozens of tick boxes plus linked ‘please tell us what…’ type of questions. Making the switch to fewer questions has the benefit of finding out what the customer thinks is important as well as reducing the length of the survey, greatly improving the likelihood that customers will actually complete it. As you say, if customers are going to be asked about their experience with ever more organisations who they come into contact with each day, the survey mechanism must be short and sweet, otherwise fatigue and reduced response rates will set in across the board very quickly.

    The second factor that will reduce response rates is if the customer feels that nothing is being done with their valuable feedback. Closing the loop and making it obvious to customers that things have been done in response to their input is vital – whether this is one-to-one communication and addressing that person’s individual issues, or some kind of broad scale action, with point of sale material “You asked for… so here it is” or other broadcast communication.

    Neil Hartley’s response made an interesting point about categorisation of unstructured text, versus non-categorised. There are companies such as Autonomy and others that already approach this from a pure pattern analysis methodology, so no pre-categorisation is done. However, our experience is that given the multitude of ways customers have of saying the same thing, categorisation is actually the best way to approach this. Our business – Feedback Ferret ( which I would be most grateful if Bob would add to his list of vendors – uses categorisation to ensure that we can avoid what would otherwise be a disintegration and dilution of particular topics. This has required many years of investment into the categorisation Lexicons – particularly in the bad grammar and bad spelling used by customers, research agents and call centres when gathering the data. With Feedback Ferret, we have components in the analysis tools to identify the unknown unknowns, evern though they may not be existing core categories, so categorisation need not preclude that type of discovery.

    Finally, whilst the exploration of feedback topics is a key part of any VoC programme, we have found that the greater part is the productionisation of the process, so that the company, its front line staff and agents can get up-to-date information quickly about the key themes and topics that affect their customers. The productionisation of the gathering, analysis and reporting is a very different process to the initial exploration and discovery of what people are saying, so both these aspects need to be considered when embarking on a VoC programme

    Piers Alington
    Feedback Ferret Ltd (United Kingdom)
    [email protected]


Please use comments to add value to the discussion. Maximum one link to an educational blog post or article. We will NOT PUBLISH brief comments like "good post," comments that mainly promote links, or comments with links to companies, products, or services.

Please enter your comment!
Please enter your name here