Human or IVR: Not an Either/Or Question

Ann All
Slide Show

Disaster Readiness and the Contact Center

Factors to consider when evaluating a contact center outsourcing provider.

After reading a Wall Street Journal article that described how some companies swapped impersonal-sounding voices on their interactive voice response (IVR) systems for more pleasant ones, I concluded such a tweak was worth a try if it was part of a larger effort to improve IVR usability. At least one of the companies mentioned in the article, insurer Asurion, seemed to get this, as the story mentioned that the company also rewrote its IVR scripts to offer an experience akin to what customers would get from a live agent.

 

I then saw two pieces that suggested that companies should abandon IVR systems and use humans to answer all calls. Writing for Harvard Business Review, Bill Taylor, author of "Practically Radical: Not-So-Crazy Ways to Transform Your Company, Shake Up Your Industry, and Challenge Yourself, " asked:

Why shove sterile technology between you and the people with whom you do business, even if you try to make that technology less sterile by infusing it with gentler tones?

He offered Zappos and ING Direct as two examples of companies that use a human-only approach in contact centers. Both Zappos CEO Tony Hsieh and ING Direct COO Jim Kelly say this kind of personal attention leads customers to sing the praises of their brands and is thus worth the added cost.

 

Tripp Babbitt struck a similar note on Customer Management, writing:

Anything that separates a customer from getting an answer quickly and accurately is an annoyance. Furthermore, it adds to costs and loses customers.


Babbitt suggested redesigning the workload so agents can respond to any customer issue. I think that should be a given if calls are going to be answered by humans-although all too often it's not. After all, isn't being routed from agent to agent, repeating your problem every time, just as frustrating as navigating through an endless IVR menu? And employees actually empowered to help customers are happier and stay on the job longer, as American Express' experience with its contact center employeesshows. Babbitt said doing this removes the need for an IVR system.

 

Hmmm. I certainly understand the point Taylor and Babbitt are making. And yet I don't think it makes sense to have agents field calls about things like store hours or account balance inquiries. I prefer using self-service channels like the Web or IVR for those types of inquiries, and I don't think I'm alone judging by a reader comment left on Babbitt's post. Wrote KOConnor1:

... I don't want to wait in queue to talk to an agent for a simple task that I can accomplish myself right away with the help of an IVR (e.g. credit card activation). I don't want to take time off my work day to wait in queue to talk to an agent for a simple task that I could easily do on my own after hours with the help of an IVR. ... I don't feel that it is possible to employ live agents for everything. ... I think that companies that are committed to customer service will shine using a combination of IVR and live agents because they will seek to understand their customers and figure out what is the best way to serve them.

Exactly! Maybe I'm a bit prejudiced by my last gig, which was covering ATMs and other financial services technology. Just as it doesn't make sense to keep bank branches open all night, I can''t imagine companies making enough staff available to quickly respond to calls around the clock. Just as using an ATM is often more convenient than going into a branch, using a well-designed self-service channel like a Web site or IVR is often quicker and more convenient than speaking to an agent.

 

I also get the point that Karen_Tiede, a reader commenting on Taylor's piece in Harvard Business Review is making, that human agents can cross-sell and up-sell products to callers, something an IVR can't do (at least not well). Dick Hunter, the former VP of global consumer support services at Dell, mentioned cross-selling as an opportunity for contact centers in a 2009 webinar on Dell's efforts to improve contact center satisfaction. To help call operations pay for themselves, Dell encouraged its agents to take cross-selling opportunities. An agent might mention that he or she noticed a customer's laptop was 3 years old, an age at which it makes sense to consider buying a battery, for example. Said Hunter:

If you've solved a customer's problem, then you've just earned a lot of credibility with them. What better time to try to make a sale?

Banks had a similar issue, worrying that online banking and ATMs removed the need for customers to visit branches, which is where they felt customers could be sold on more lucrative products like loans and investment services. In reality, they found customers were more likely to respond to product offers on Web sites and request a follow-up contact from a human. Most bank tellers don't have the time, inclination or training to cross-sell, so cross-selling opportunities at branches were mostly restricted to signage.

 

Some banks had success with a concierge concept, where all customers were greeted at the door and briefly asked about their banking needs instead of just shuffling off to a teller line. Harvard Business Review reader Richard Goh suggested a similar option for IVR. He wrote:

... I think that the routing role should be done by an actual human, who can listen and divert to the right personnel for help, rather then a menu that takes 2 minutes to read out and then goes into another 2 minutes in the next level. Its a feeling of actually making contact that matters. Further, a menu rather than an actual human being gives a feeling that you have not made contact yet. Proximity is not the distance but the time it takes to get there. How many companies can say they have true proximity with their customers?

Another reader, Gnomic, hit on a point that neither Babbitt nor Taylor address: Some customers will call with low-value, time-wasting inquiries that will keep agents from serving other customers. He wrote:

Picture a company with 50M customers. 30M don't ever call in. 25M call in occasionally and are either happy or not unsatisfied. 5M call in a lot and 1M call in everyday and almost always talk to someone, sometime multiple times a day. As a business person, I'd look closely at that 1M and pick out the 1K customers worth keeping and give them a private number or otherwise route them to special servicing. The rest I'd encourage to move to my competitors. Why? To drive up their costs, not mine. I might even make them a special offer to move.

When I interviewed Lior Arussy, founder of the Strativity Group, he told me that's exactly what ING Direct, one of the companies lauded by Taylor in his HBR piece, does, "firing" some of its unprofitable customers every year. He explained:

These people call the call center too much. The bank tells them, 'I cannot do all of this hand-holding and still give you these attractive rates.' Ultimately, it will cost the other customers. That's the discipline that is missing in order to start managing customer relationships in a profitable way.


Add Comment      Leave a comment on this blog post
Nov 5, 2010 2:28 AM Kirsi O'Connor Kirsi O'Connor  says:

Thanks for a great and well-researched article, Ann! I found the information on Gnomic and ING "firing" the most unprofitable customers interesting, just something that I had never thought of in this context. Earlier this week my mother-in-law told me about an illness-ridden friend who was booted off my her doctor, apparently "randomly" but most likely for requiring too much time and effort. Neither the friend, my mother-in-law, I or any of the tens of other people that heard the story were impressed. 

Reply
Nov 5, 2010 12:36 PM Tripp Babbitt Tripp Babbitt  says:

Ann-

Let's be clear.  Shifting workloads is neither what I said or a good solution.  What I said is the work should be REDESIGNED to eliminate failure demand (demand caused by a failure to do something or do something right for a customer) which represents typically 40-60% of all customer demands in most contact centers.

IT people that fail to redesign the system to optimize effectiveness miss opportunities for improvement.  They accept the failure demand and automate it . . . this helps no one.

My bank spent millions over a year ago on CRM technology and they still do not know who I am.  I suspect that some of it is because it is all too impersonal from a relationship standpoint.

As for demands about when the bank branch is open, the question becomes what is the demand.  Do they need to come into the bank?  Could you have saved them a trip by giving them a competent agent to talk to? 

Technology folks need to understand that standardization of menu options on an IVR does not absorb the variety that customers bring to it.  Study customer purpose and demands first before any technology is put into any service organization.  My findings have been that you will have better relationships and happier customers.

Oh and for those bothersome queues are created by all the failure demand and poorly designed work systems.  If demand is predictable after reducing failure demand you should be able to handle all calls.  For those unpredictable ones I often find they are man (and usually) marketing made.  Some promo to offer something and no tells the contact center . . . the result they get slammed.  The ones that come from freak errors or events are going to have long queues anyway. Service Organizations will need to deal with the demand eventually (re-booking flights after a weather event is an example) you will need to deal with the demand eventually.

There are some things that are best done by humans and IT should be pulled if it will enhance the work.  But understand the work first and be able to study it and redesign it before entrapping IT is a solution.

Reply
Nov 6, 2010 7:24 AM Ahmed Bouzid Ahmed Bouzid  says: in response to Tripp Babbitt

The notion that "Technology folks" would be designing these systems tells you everything that is wrong with most IVR systems out there.  Designing, deploying, maintaining and tuning a highly usable IVR system is not an IT project, just as deploying a highly effective web site is not an Engineering task -- and we all know how websites developed by engineers or non designers look like.  They look nasty bad....   Designing Voice User Interface (VUI) systems is a far more challenging endeavor than designing a GUI (website), and the number of people who know how to design systems that people would not only tolerate but actually enjoy using is a small one indeed.   So, you can see why IVRs are more often than not pretty awful....

Reply
Nov 8, 2010 2:41 AM Tripp Babbitt Tripp Babbitt  says: in response to Ahmed Bouzid

Ahmed-

Thanks for pointing this out.  I did not intend for this to be just "an IT project."  But that IT folks need to understand demand and purpose before embarking and this should involve both management and agents. 

Reply
Nov 15, 2010 7:27 AM Laura Chumley Laura Chumley  says: in response to Ahmed Bouzid

Right you are, Ahmed. IVR's are all too often spec'd by stakeholders with no grasp of VUI design elements, or knowledge of what creates a smooth, effective interface. Why do companies investing large amounts of money in these applications accept this, do you suppose?

Reply
Oct 13, 2011 8:52 AM Blaine at BPO Industry Blaine at BPO Industry  says: in response to Laura Chumley

Laura, I don't think it's as black and white as you think. Companies invest large amounts of money into these types of applications because they can change the way IT communicates.

Reply

Post a comment

 

 

 

 


(Maximum characters: 1200). You have 1200 characters left.

 

 

Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.


 
Resource centers

Business Intelligence

Business performance information for strategic and operational decision-making

SOA

SOA uses interoperable services grouped around business processes to ease data integration

Data Warehousing

Data warehousing helps companies make sense of their operational data