Survey Says… Not Enough Information


Share on LinkedIn


Tried and True. Or is It?

Those of you following my blog posts know I travel a lot. I’m that guy – the one that gets on the plane before everybody else because of my frequent flyer status. Last night, I had to make significant changes to my travel plans – the changes were so significant that I had to make a call, versus managing them online. I won’t tell you the name of the airline, but I will tell you I was on the phone for about 45 minutes. One representative handled the maze of airline policies and procedures, but had to consult five other people to resolve my request. At the end, I was asked to take “the survey.” I answered all of the questions with focus, all of them directly about the specific agent that helped me. She did a great job on my behalf – really. The facts are that: I had to be on the phone for 45 minutes while she talked to five people to do something relatively simple, Next, the airline thought it would be okay to ask me to pay some $600 in change fees that were never mentioned in the survey. I am disgusted by this airline (as I write in the air, on the flight I changed). I plan to build “status” with one of their competitors during next year’s miles thronging. But, they don’t know it, because they weren’t listening to the right information.

The art of the survey has never really gone away, especially in the realm of the contact center. It remains one of the most popular ways to gain feedback from customers. Maybe it comes in the form of an interactive voice response (IVR) following a call, or you’re told after you make a purchase to call the number at the bottom of a receipt, or you receive a text after you activate a new product or engage in tech support. However, the general format of question/answer is still followed and the goal is still to gauge customer satisfaction. But after my recent encounter, it really begs the question of how effective they are at understanding the customer’s experience or accurately indicating loyalty.

Experience Matters

The trouble is, customer experience reigns supreme, so companies can’t afford to be left in the dark about how they stack up. According to Accenture, “price is not the main reason for customer churn, it is actually due to the overall poor quality of customer service.” RightNow’s Customer Experience Impact Report backs this up by stating that their research indicates that 89% of consumers began doing business with a competitor following a poor customer experience. So it only makes sense that companies invest significant resources into finding ways to gauge customer satisfaction. They’ve created Voice of the Customer programs, of which surveys are typically at the center, and use a variety of analytic tools to measure results.

Incomplete Picture

But surveys are all measuring different things, and often times different areas of the company are working independently of each other to collect different data. Gathering the results takes time, but more importantly, the results often don’t give indications of what really happened to create a positive or negative experience. Companies need to understand a customer’s collective interaction with their organization, and not have information siloed by departments or be limited to the information that can be collected in structured data form.

To understand what formed an experience, not just an opinion, you often need more than a survey can provide, especially if it’s composed of only structured questions which do little to tie events to outcomes. Companies often think they’re collecting a sufficient amount of information, especially if they’re incorporating additional data points such as buying patterns and demographic information, but that still leaves a lot of gaps in the story. Throw in the fact that surveys are based on a very small percentage of your customers, and contain vague answer options like “somewhat satisfied,” and companies are very challenged to make informed decisions based on their results.

IA image 1

The Missing Piece

The largest untapped asset companies have to help uncover what’s causing negative experiences for their customers are the conversations captured by their contact centers. Using a speech analytics or interaction analytics technology to put structure to the audio, organizations can now find the specific events that elicited confusion, frustration, anger, dissatisfaction, etc. Taking this a step further, companies can quantify the number of times each of these specific events occurred to determine their impact across the entire business. This allows them to determine which business processes and agents behaviors are directly affecting their customer satisfaction scores, so that corrective measures can be put in place. This is far more valuable than simply receiving survey results that, through 2-3% of your customer interactions, statistically tell you that only 38% of customers would be likely to repurchase from you based on their last interaction. While that may be interesting, it’s not actionable. This dated approach gives no indication as to what happened during the last interaction to cause that result. And, other than with the 2-3% of the customers surveyed, there is no way to actively manage repurchase at the individual level. Companies need specific context, at an individual call/customer level, that gives an unbiased account of what happened and why in order to be able to determine the ramifications of the interaction. Customer experience happens at the individual level, and companies are most productive when resolution is done in the same way.

A Step Ahead

The most progressive companies are taking a “conversational big data” approach to customer satisfaction and combining the information contained in the audio with traditional structured data to build predictive models. By studying the events that took place in customer journeys with known outcomes, correlations and trends are uncovered that allow businesses to predict churn, loyalty, and CSAT scores – all without ever having to issue a survey. This information is based on 100% of the interactions, eliminates departmental silos, and provides a holistic understanding of what’s driving customer sentiment and behavior. Every interaction, and every event that occurs in that interaction, becomes the “survey.” Until recently the technology necessary to make this happen did not exist.

IA image 2

Final Thought

Surveys will always have their place and will no doubt continue to be a valuable tool. However, interaction analytics offers companies a way to use the information they’re already capturing in their contact center to affect transformational change. Through the creation of metrics, the events that impact customer experience are not only uncovered, but more importantly, measured, so that their reach can be understood. This affords companies the opportunity to quickly address issues and measure the effectiveness of their efforts. And in the era of the fickle consumer who is ready to trade in their loyalty at the drop of a hat, how can you afford not stay on top of your brand experience?

{Photo Credit: Wikimedia Commons}

Republished with author's permission from original post.

Ryan Pellet
Ryan is SVP at Nexidia - the leader in interaction analytics. He responsible for leading Nexidia's Strategy and Services where he creates customer strategies for the Fortune 100. He is: a recognized expert in using behavioral / business analytics, author, keynote. Prior, Pellet was a VP at Convergys. There, he led both the Center for Applied Customer Analytics, and Global Consulting – using CVG's 2 billion+ customer interactions. A serial entrepreneur, Ryan founded the Finali Group, which was acquired by Convergys. He also was an executive/founding contributor to Accenture's CRM Practice.


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