If you’ve sat in on any of my webinars on First Call Resolution (FCR), you’ll know that what’s important isn’t what the definition of FCR is, but rather that you calculate it consistently across the board and that it gives you some meaningful information.

When I started doing some research into the actual definition of FCR used in the industry, it became pretty clear that the world is split into two camps.

Camp 1:                The total number of calls (or contacts) resolved right the first time divided by the total number of calls.

Camp 2:                The total number of calls resolved right the first time divided by the total number of first calls.

These give you the same basic information, presented two different ways.  Camp 2 is a better number if you are measuring FCR through surveys, since you won’t actually have the total number of calls available (since the survey isn’t a 100% sample size) . However,  on its own it doesn’t take into account  any non-resolved calls.  If you have a lot of unresolved calls, they don’t show up. For example:

• 68‘One and Done Calls’
• 100 ‘Issues’
• 200 follow up repeat calls (extraneous information)
• FCR⁽²⁾ = 68/100 = 68%

If you’re in the first Camp, then you’ve got the total number of issues and the total number of calls:

• 68‘One and Done Calls’
• 100 ‘Issues’
• 200 follow up repeat calls
• FCR⁽¹⁾ = 68/300 = 23%

In Camp 1, your FCR number is also intrinsically related to how well you resolve calls (average number of calls to resolve), because as the total number of calls drop, your FCR gets better:

• 68 ‘One and Done Calls’
• 100 ‘Issues’
• 50 follow up repeat calls
• FCR⁽¹⁾ = 68/150 = 45%

So when a number of 68% FCR is bandied about as the industry average, I must confess my confusion.   If this is calculated as FCR⁽¹⁾ then it’s a totally different number than when calculated as FCR⁽²⁾.

Ah well. Statistics; gotta love ‘em.

I’m putting part two on the Average Call series on a short hold. Came across something while reading my morning paper on Tuesday.

On the front page of the Life section, there was an article titled “Got a complaint? Post it on social media for a quick response”. It was a clarion call for customers to bypass the queue, and go social to get what they want. Quoting from the article, “Didn’t receive dipping sauce with your order? Ignored by a salesperson? Whine on Facebook or Twitter, and you’re guaranteed an instant response from the company’s social media rep.”

The piece centered on one Virginia Sokoloff (no relation) who had had her outbound flight canceled. She had issues at the airport getting a new outbound flight, and in the process the airline desk agent voided her return ticket. It was only on her return that she discovered this, and had to buy an expensive one way ticket to get home. Understandably, she was upset and wanted the company to make it right.

So, she posted on the company’s Facebook wall, received an apology, a ticket refund, and a discount on her next flight.

After being ignored by 3 agents at her home airport, getting no help at the return airport, and receiving no satisfaction through customer service email and phone.

Great. We now have mass media encouraging customers to do the ultimate zero out of queue, where they can get someone that can actually help them while publicly embarrassing the company.

With 51% of company Facebook interactions negative (a stat from the aforementioned article), it’s clear that social media is becoming a broadcast channel for poor customer service. The recommended approach to handling social media is to make sure that there’s always someone from the company listening. But the more successful that social media channels become for customer resolution, the more that customers, deserving and non-deserving will start there. That 51% is only going to go up, and the reactive approach will break down because it won’t scale.

I respectfully submit that the real effect of social media will be that customer service will be forced to meet a higher standard. For everybody, not just your gold customers.

One last thought – if you have to provide an avenue for a customer to escalate to someone who can really help, isn’t there a less public way of doing that?

(Postscript- I’ve intentionally have left out the airline name in the posting. I use them frequently, and have found that their service has dramatically improved over the last five years, to the point where they are my preferred carrier. Nevertheless, it can’t be denied that what happened to Ms. Sokoloff was a class one major fail.)

I deal with so many different agent desktop applications in the day to day of dealing with different customers, prospects and industry professionals.  I’m truly amazed at the variation of what has been implemented in contact centers across North America – from the state of the art, integrated desktops that provide the agent with a single cockpit at a glance, down to what still remains the number one application in use in contact centers today (from our experience) – that workhorse, the AS/400 green screen.

When companies look at updating their agent desktop applications, they evaluate vendors for what they are considering to be a commodity item, and end up comparing feature to feature on the chart, and trading that off against the cost of the system.

What often gets lost by both the business and the technology groups is the fact that, whatever the choice, this is an application that has to able to provide meaningful value to the agents, and do it in such a way that it’s convenient for them to use 60 to 80 times a day, as they deal with call after call after call.

Flipping back and forth between web pages, moving the mouse around in a scrolled screen, or duplicating fields are all things that agents hate; they are things that cost you time and create errors – and they are all things that agents will skip if they can.

Remember that agents are conscious of handle times, so they still do want to be quick.  More importantly, if they don’t see a ‘What’s in it for me’ (WIIFM), they will skip what they can. Worse, they will see it as just another BOHICA (you’ll have to look that one up yourself!).  And you, as a management team, either deal with a sub-par, underutilized system, or worse –  deal with the fact that you purchased a very expensive enterprise application that gets minimized and ignored every morning.

Keep your agents in mind.  Over and over and over again.

Measure the time for every call.  Add ‘em, then divide by the number of calls.  Presto, “Average Handle Time!”

The number one measurement for call centers.  Why?  For that, we can thank (or blame) a Danish telephone engineer named A.K. Erlang.  He was trying to find out how to determine the number of expensive telephone circuits needed to handle the calls of the (presumably Danish) population.  It involves a bunch of math that I used to be really good at years ago.

For call centers, if you substitute agents for circuits, Erlang tells you how many expensive agents you’ll need to handle the calls, based on an average call length. So, if you know AHT, you know what it’s going to cost you.  And if you get the call times down, you’ll save money.

Simple.  Powerful.  And…  dangerous?   Why?

In most contact centers, effectiveness measurements (FCR, CSAT, Quality), are derived through a subjective process on a sampling of calls.   But, AHT is the one thing all call centers can empirically measure, based on 100% of the calls.  Thus, it’s the one measurement that is undisputable, and is believed to be actionable.

Stated another way, for efficiency, we have a measurement that we can get instantly and we trust.  For effectiveness, we have a measurement that we get long after the original event, and is, at best a guess.  Which one gets used to make decisions?

And therin lies the danger.  In my next post, we’ll examine how potentially dangerous this can be.

A topic that frequently comes up between senior call center people is that their voice is all too rarely considered by their senior executives.

Last fall, I heard a great keynote from a call center VP where she talked about the company’s core values of customer satisfaction and loyalty.   And how, by managing to those values she and her team were able to transform the operation.

But, it was in a session when her peers were discussing ways of making themselves more relevant to the executives that struck a chord with me.  To paraphrase (clumsily), she said that she would go into weekly meetings fully armed with handle time and occupancy numbers, and watch the execs visibly disengage.

So, she decided NOT to do that anymore.  Didn’t ask anyone, just stopped giving them those stats.  And, here’s the really surprising thing.  No one said anything. Probably no one even noticed.

Instead, she started discussing customer satisfaction and customer feedback.  Topics related to the core company values and to the executive goals (and probably bonuses).

And they started paying attention.  They started seeking her views and guidance on customer issues.

Makes sense to me.

Speech analytics can be broken down into two distinct components.  The first is the speech recognition and conversion of audio to text.  The second is basic text analytics or semantic analysis that can be run on the output text files.  Even speech analytics applications that run in “real time” go through this two step process in some form.

Speech recognition technology works best when trained for a specific user.  In a contact center environment, accuracy is reduced because the users are anonymous, and accuracy is further reduced when accents are introduced.  The chart below looks at accuracy – but it stops almost ten years ago.  I assume that the speech recognition has made significant progress since then (“Switchboard” recognition shown).

From http://www.itl.nist.gov/iad/mig/publications/ASRhistory/index.html

Nuance probably has the most advanced speech recognition software currently available, and claims rates of 95% accuracy (still below human error as per the chart above).  However, Scientific  American noted in December of 2010 that for dictation software (that is, trained for your specific voice and accent) “the software isn’t much good unless you are speaking into a microphone, without background noise, preferably without an accent. You still have to speak all punctuation (“comma”), like this (“period”). “(http://www.scientificamerican.com/article.cfm?id=talk-to-the-machine)

I understand the value of Speech Analysis in a call center environment, where it can drastically reduce times for locating speech segments, or key off of pre-programmed specific words or phrases.  But I keep hearing that people are using it to determine the reason for the call in the first place.  I’m honestly confused as to how this can work.  And if you can determine the intent of the caller from analyzing the voice recording, shouldn’t you build that into an IVR?  Maybe Interactive Voice response needs speech analytics.  Oops – did I just come up with a game changing idea?

Every year we attend a couple of major conferences, where, as a sponsor, we are asked to come up with a subject that’s forward looking, new and exciting for the contact center industry.  It’s often a frustrating task, when I consider what I call ‘the speed of business’. I think IBM had an ad back in the ‘90’s about ‘moving at the speed of business’.  But from everything I see, that’s glacially slow.

For most established companies, the days of ‘just do it’ are gone.  Decisions are committee and consensus based, and as result, new technology is slow in evaluation, slow in approval, and slow in implementation.

Consider call center email.  This is not a new, exciting technology. It’s different from just ‘email’ because it involves routing to the next available agent based on a skill set and properly tracking their work efforts.   This is stuff that’s been around for at least 15 years, in much the same form as it is today.  And as a company that provides email technology, I’ve seen directly that the uptake on installing email software to properly manage customer contacts and agents has only really hit the mainstream in the last couple of years.

I wonder how long it took people to believe that they should install phones back in the early 1900’s?

So those exciting technologies that we need to talk about at conferences are cool and exciting, but they are usually ten years away from adoption and don’t really drive business decisions today.

While many companies are keenly aware of the value of excellent customer service, few can accurately quantify its direct impact on their bottom line. Most businesses understand that turning the customer experience into an emotional engagement strengthens their brand and results in more loyal and satisfied consumers. However, most companies do not measure their effectiveness in converting each opportunity into revenue, nor do they measure the overall cost of poor customer service.

According to a recent survey, poor customer service costs $338.5B per year in lost business. The hardest hit industries across all countries surveyed are financial services, cable and satellite TV providers, and a variety of telecommunications companies.

Why do customers leave?

According to the survey, consumers are remarkably consistent in citing key reasons that they leave. Assisted service is well developed, with the overwhelming majority of consumers saying their most satisfying experience occurred because of a capable and competent customer service representative. But self-service that is not intelligently integrated with assisted service is a key area of concern. Consumers feel the most significant root causes of poor service are:

• Being trapped in automated self-service
• Being forced to wait too long for service
• Repeating themselves
• Representatives that lack the skills to answer their inquiry

What would make them stay?

Consumers are demanding better integration between self-service and assisted service, including voice self-service and eServices. The most requested improvements in all countries were to be able to start in voice self-service or the Web and get live assistance from an agent, and to start in e-mail and have better integration with agent-assisted service. In other words, “Don’t ask me twice.”

In addition, consumers said that the ability to communicate across multiple channels is critical to loyalty. When asked what they would most like to see companies deploy to improve service, 40% chose human service, but more than half of consumers chose at least one new communication channel among their top choices. In other words, “Treat my interactions as a conversation.”

Finally, overwhelmingly, an average of 86% of consumers per country said they would find proactive engagement either a “strong benefit” or would “welcome proactive assistance” when they were stuck on the Web or in self-service. In other words, “It’s ok to give me a call or ask me if I need help.”

The Bottom Line

Poor customer service has a clear and immediate impact on a company.  Companies need to understand and measure the direct business impact of customer service, and identify the gaps between the customer experience and expectations. To do so, companies need to assess their existing communication channels, and make interactions more convenient across multiple channels. It is imperative that an enterprise engage consumers on the customer’s terms or risk losing them.

As I get the opportunity to talk with many people that manage call centers, I find it fascinating to see two distinct views on the capabilities of agents.

First view – Agents are Angels.  They are trustworthy, and care about the customer.

Second view – Agents are Demons.  They cannot be trusted, and they are putting in time for a paycheck.

I can pretty well identify which view will be put forth by asking ONE question.  That question is, “What role does Average Handle Time play in your call center?”

The call center that makes AHT the number one goal will almost certainly view their agents as Demons.  Because, by placing call duration above doing the right thing for the customer, the Agents are effectively being encouraged to do bad things:

Going to be a tough, long call?  Better transfer it!

Don’t have the information at hand?  Make it up!

Not mandatory entry?  Then don’t enter it into the CRM!

The message is “Don’t do ANYTHING you don’t absolutely have to that takes time.”

Clearly, this is not what the business really wants.  These actions tick off customers and deprive the business of a true picture of what’s going on.

Now, don’t get me wrong – AHT is important.  I’ll explore this further in upcoming posts.

I’ve always thought of reading blogs more like a pub crawl than a guided tour.  Read a bit, learn a bit, and stumble off to somewhere else for another sip.  Much of the power comes from establishing links between a blog’s content and other valuable information.  We believe that by providing useful information that these links will happen over time, and that patience is a virtue here.

Others, listening to the search engine optimizer folks are looking for a short cut.  While we encourage responses to the articles, they ARE moderated in order to filter out these link vultures.  For your amusement, I thought I would share some of the recent blog responses that didn’t make the cut:

“This is very interesting, You’re an overly skilled blogger!”   (A Russian diet site)

Remind me to cut back on my skill.  Don’t want to overdo it.

“I am really satisfied with this posting that you have given us. This is really a stupendous work done by you.”  (A financial management site).

I’m pleased my stupendous posting left you satisfied.  But, I’m still holding on to my wallet REAAALLY tight.

“I’m gonna watch out for Brussels.”  (Watch free TV online)

Sneaky, sneaky Brussels.  It’s always best to keep an eye out for Brussels.

“I enjoy, result in I discovered exactly what I was having a look for. You’ve ended my 4 day long hunt!’ (Dating site)

Four whole days of hunting!  For what I had to say!  Glad you enjoy!

“Some really superb info.  Sword lily I found this.” (No idea, except the word ‘insane’ is in the name)

I’m thinking that “Sword Lily” is an exclamation, like “Gadzooks!” or “Zounds!”  Or not.  Could be the insanity talking.

Anyways, we encourage you to share your (real) thoughts.