Interactive Intelligence Blog - Capturing the Voice of the Customer: Part 2 -- How to do it

Capturing the Voice of the Customer: Part 2 -- How to do it

Posted on 17 Aug 2009 by Joe Staples

Joe Staples

Last week I blogged about the need for businesses to capture the voice of the customer. See:
http://www.inin.com/Blog/archive/2009/08/04/Capturing-the-Voice-of-the-Customer-Part-1-The-Need.aspx I'll assume that you all left that riveting post, fully convinced. This "part 2" piece looks at the best way(s) of actually accomplishing the task of getting your customers to rate your products or services.

I think you've got five main options for capturing the information: mail, phone interviews, email, automated web survey and automated post-call survey. The chart below looks at three dynamics of each option: the response rate, the cost, and the timeliness. Each has its upside and downside, but mail, phone, and email all have reasons to be ruled out as choices. Unfortunately for most companies, use of one of these three methods is the norm...IF (a big IF) they are actually doing any surveying at all.

So what is the ideal method? Supported by the above chart, it is automated post-call surveys or web surveys. And I'd make the case that if the interaction is a phone interaction, that a post-call survey at the conclusion of the call is by far the best method. If your service was provided via the web, that a web survey is the best method. Back to the basic rule of communicating with the customer the way THEY CHOOSE to communicate with you.

For both automated web surveys and post-call surveys, the technology is available and it works very well. Just a side note here to explain an "automated post-call survey"....following the call into your service center, the caller is automatically transferred to an automated system where he/she is prompted with a few pre-recorded questions. The caller responds with DTMF input and recorded responses. The data is aggregated and scored. Dissatisfied customers or prospects are immediately escalated. Pretty straightforward, but I wanted to be sure you understood the process.

Here are some guidelines to consider regarding implementation of technology to capture the voice of the customer:

Have the customer opt-in to the survey up front, before the service is delivered. Otherwise, you'll only tend to capture those polar experiences of either great or terrible service.
The call center agent can't have any influence regarding who gets surveyed. They can't be required to transfer the caller into the survey application. Otherwise, you won't get any of the dissatisfied customers in your survey mix (hey, it's the reality).
For post-call survey technology, the application is most useful if it is provided by the same vendor that provides the ACD and call recording system. This let's you correlate the call, the recording, and the survey.
Don't make your survey too long. Three or four questions is best.
Ask questions on different subject areas -- i.e. courtesy of the agent; did the customer get the answer he/she needed; timeliness of the response; etc.
Implement a system for immediate response for any survey where the customer gives you a poor grade. As the chart above indicates, web and post-call survey technology gives you realtime response capabilities. Imagine the reaction of a customer who gives you poor marks and within minutes receives a live phone call from a supervisor wanting to get information and working to resolve the dissatisfaction. You just got a huge benefit over the companies who can't react quickly and simply just pay the price of customers deserting them and making their next purchase from a competitor.

So that is about it. Not too complex and pretty darn cost effective ways to capture the voice of the customer and use it as a competitive advantage.

What has your experience been? Anything you'd add to the my list of six implementation tips? Please comment for others to learn from you.

Joe Staples -- customer voice evangelist and drive time blogger

Enterprise-wide communication applications in an OCS world

Watch demo

Using your communications system to automate core business processes

Download whitepaper

Third party study outlines successful migration strategies

Download report

Tags: Best Practices, Contact Center, Market Trends and News

Share this post:

Comments

Ian Jacobs commented on Monday, 17-Aug-2009

Good analysis Joe.

One point I'd add: if enterprises are doing post-call (or post contact, if multi-channel) surveying, the results for call/contact resolution may not be accurate. Customers often don't know if the interaction solved their issues immediately following its completion -- it sometimes takes a few days with no further issues before the customer feels that the call/contact did the trick.

Joe Staples commented on Monday, 17-Aug-2009

Good point Ian. Different than the question, "was the agent courteous?" I suppose that on those items that might not have a resolution yet (i.e. "Was the issue resolved to your satisfaction?"), it would be a best practice to include an option for "too early to be determined." Thanks for your comment!

Joe commented on Tuesday, 18-Aug-2009

Could you clarify #4? - Don't make your survey too long. Three or four questions is best.

In I3's feedback class we learned about ACSI compliant surveys. In order to be compliant you have to have more questions than that. Shouldn't we promote ACSI compliant surveys?

Joe Staples commented on Tuesday, 18-Aug-2009

To item #4 (and thanks for the question)...my point is that in the design of the survey, you need to not make it too long or too complex. Otherwise, you'll watch people drop-off or opt-out. That said, my 3-4 question suggestion might be a little too strict. The ACSI model (American Customer Satisfaction Index)is an outstanding method for defining how to effectively measure customer satisfaction, and in-turn how improvements in customer satisfaction affect a company's bottom line (see: http://www.cfigroup.com/acsi/overview.asp). And the CFI group (patent holder to the ACSI methodology) is much more of an expert on satisfaction measurement instrument design than I am. While I do know that they are of the opinion that asking too many questions is not a good practice, I would defer to them as a good resource relative to the actual guidelines for designing your survey. Thanks again for the comment.

Ben Engelsberg commented on Thursday, 20-Aug-2009

Joe,

Great pair of articles. I really wish we had something like the ability to conduct automated after call surveys at my last job, it would have made capturing the voice of the customer much simpler.

You have called out the importance of escalating dissatisfied customers to supervisory staff, to remediate their dissatisfaction, and collect more detailed information. I also believe that it is important to conduct more detailed interviews with (some of) your most satisfied customer, in order to capture detailed information about what makes your call center stand out, so that you can identify, retain and enhance the things your call center is doing especially well.

Gina Clarkin commented on Friday, 21-Aug-2009

Ben,

Great point on following up with your most satisfied customers. The next obvious question is: How do you know who your most satisfied customers are? If the survey application is provided by the same vendor that provides the contact center platform (as Joe suggests), there is no need to piece together data from multiple disparate systems. The system already knows everything about that customer's interaction that may have impacted the satisfaction score - not only who the agent is who handled the call, but also call statistics, the path of the call, what was said, and who the customer is. Identifying the right customers for follow-up becomes simpler, and interviewers are equipped with better information about the interaction.

Chrystele Maire commented on Monday, 24-Aug-2009

For automated post-call surveys, cost per survey is shown as low. However it would require some upfront investments with the ACD provider as this functionality is usually not provided by default, wouldn't it?
Advantage of email or web surveys is that technology is usually there, ready to be used.

Joe Staples commented on Tuesday, 25-Aug-2009

Chrystele, You bring up a good point worth considering. My "low cost" observation was relative to the ongoing costs of admistering the surveys and capturing the data. You are right that there are up front tech costs associated with the application, but I believe the ROI would be strong since once the application is acquired, the ongoing costs are near zero. Add to that the timeliness of the information and I think automated post-call satisfaction surveys are a very practical, cost-effective approach to capturing the voice of the customer

Ismael Jiménez commented on Tuesday, 25-Aug-2009

The only shortcoming I see with post-call survey is that it only applies to get the satisfaction rate for the interaction itself (and mainly limited to calls). What about assessing the satisfaction about the core service the company offers (as it's the "core business", that's precisely what matters the most). Think about assessing the satisfaction in processing a mortgage request (in banking), fixing a leak in my home water pipe (insurances companies) or activating my new high-speed Internet Connection (telecom providers). This business processes usually start with a phone call to the company Contact Center, but doesn't necesarily end up the same. Sure i will more than willing to evaluate how polite was this agent when he promptly entered my request in their business process system, but it's more important how satisfied i'll get once the entire process will be carried out. I don't clearly see how post call surveys can help here...

Joe Staples commented on Wednesday, 26-Aug-2009

Ismael, the same technology (and associated benefits) can be used for "post-service" satisfaction surveys. The difference is the survey is initiated with an automated outbound call at the time the service is completed (i.e. a multi-step process of having an internet connection is installed and the customer request is closed). The technology still automates the information gathering. The information is available near-realtime. Once the technology is in place, the cost of gathering the feedback/information is near zero. Your point of paying attention to overall customer satisfaction and not just measuring individual calls/interactions is a good one. As noted above, I believe the outbound, post-service automated satisfaction survey is a good method to gather that info. Thanks for commenting and sharing your views. Joe Staples

Ismael Jiménez commented on Wednesday, 26-Aug-2009

Just to round up the post call survey application, what about adding some speech analytics functionality to quickly summarize the most frequent compliments/complaints that customers might express in the form of recorded free comments?. That's pretty straighforward using web based surveys and they are already entered as text but it implies some work to be done by listening to free comments recordings and writing down the key points.

web development commented on Wednesday, 2-Dec-2009

Hey, that was interesting,

Keep up the good work,

Anyway, thanks for the post

Submit Comment

* - Denotes Required Field

Name: *
URL:
Email:
Comments:
CAPTCHA: *

Archives

Tags

Capturing the Voice of the Customer: Part 2 -- How to do it

Comments

Submit Comment