The old days of "measure everything that moves" ended with the advent of customer relationship management nearly a decade ago. Today's top contact centers realize that, to be successful, they must be able to cut through the clutter of available stats and data to find the metrics that truly impact and reveal key insights into the customer experience. Basing an entire service strategy on the number of calls handled per hour or on average handle time will inevitably lead to shortcomings in quality -- and short fuses among customers. On the other hand, focusing too strongly on quality metrics with a disregard for quantitative and efficiency measurements can still have an adverse effect on the customer experience, not to mention the center's costs.

There is no single perfect or complete list of performance metrics that fits all contact centers. However, based on shared practices of the best centers, shared advice of top consultants, and the opinion of this author, the following are the seven key metrics that are most critical in gauging and securing customer satisfaction, loyalty and contact center effectiveness.

First-Call Resolution

Recent research suggests that no single KPI has a bigger impact on customer satisfaction than does first-call resolution (FCR). According to customer contact research and consulting firm Service Quality Measurement (SQM) Group, customer satisfaction drops an average of 15% with each callback a customer must make to a contact center. SQM Group has also found that for every 1% improvement in FCR, you get a 1% improvement in customer satisfaction. In a study of more than 150 contact centers, SQM found that centers that achieved "world class" customer satisfaction ratings had a FCR average of 86%, while centers that were not among the elite in customer satisfaction had a FCR average of only 67%.

Increased customer satisfaction isn't the only big benefit realized by contact centers that achieve high FCR. These centers typically also enjoy:

• Lower operating costs. A low FCR rate breeds a high number of repeat callers -- and frequent callbacks aren't cheap in high volume contact centers.

• Reduced revenue at risk. SQM research shows that if the customer's issue is resolved on the first call, only 3% of those customers are at risk of defecting to your competitors -- compared to a whopping 38% of customers at risk of doing so if their issue isn't resolved on the first contact.

• Higher employee satisfaction. The strain on agents who must contend with frequent callbacks from often-frustrated customers is significant, and invariably leads to low morale, poor customer service and high agent turnover. However, when agents are given the tools and training they need to achieve high FCR, they feel empowered and confident on calls. Invariably, customers take notice.

While FCR is quickly emerging as a critical KPI for contact centers, there is some question about what exactly constitutes a "resolved call." Some centers consider a call resolved if the agent didn't need to transfer it. Others deem a call resolved if there is no follow-up work to complete after it.

Certainly, shooting for calls that require neither transfers nor follow-up work is a sound approach to high quality service, but it is incomplete from an FCR measurement standpoint, say experts, because it fails to take into account something essential -- the customer's perspective.

"Many contact centers have employed technology and manual solutions to help them answer their FCR rate question," says Dr. Jodie Monger, president of Customer Relationship Metrics and a pioneer in customer satisfaction measurement for contact centers. "Some of these solutions cost thousands of dollars to implement, but not one of them can answer the question better than the customer. Reviewing phone records and running a software application is the ultimate beating around the bush in this case. Stop trying to get to your FCR rate via the back door. Go to the source."

More contact centers are starting to take Dr. Monger's advice, implementing real-time customer feedback applications -- phone, IVR or online surveys that customers complete immediately after the service interaction. The best measurement method, according to Monger, involves the use of CATS (completely automated telephone surveys), where callers, while waiting in queue, are asked to participate in a post-call survey. Those who accept are automatically routed to a concise IVR survey following their interaction with an agent, and asked five to 10 focused questions about their experience, with the emphasis on whether or not the customer felt the call was fully resolved with a positive outcome.

Not only does asking customers whether or not their issues were resolved provide a clearer picture of the call center's true FCR rate, it can help the center (if the survey is appropriately designed) discover some of the main causes of repeat calls, says Monger. "For those [customers] who had a problem that was not resolved on the call, the survey should branch to an open-ended question to capture the customer's description of the problem. This qualitative information adds the explanation to the dramatic quantitative information you now have available. The cause of unresolved calls is invaluable to correcting process issues."

Service Level/Response Time

Service level and response time are classic metrics, ones that lie at the heart of effective contact center and customer management. Without service level and response time objectives in place, you wouldn't know how accessible the center is to customers, how many agents are needed to provide efficient service, or how your center's service compares to others in your industry.

Service level is defined as: "X percent of contacts answered in Y seconds," e.g., 80% of calls answered in 20 seconds. Response time (which is the equivalent of service level for transactions that don't have to be handled the moment they arrive) is defined as: "100% of contacts handled within N days/hours/minutes," e.g., all customer email inquiries will be handled within four hours.

In essence, service level and response time objectives tie the resources you need to the results you want to achieve. These metrics measure how well you are getting customer contacts in the door and into the hands of agents (and thus are essential for planning and budgeting). They are the clearest indication of what customers experience when they attempt to reach your contact center.

When establishing and assessing service level and response time objectives, the important thing isn't merely how high your overall stated objectives are, but how consistently the center hits those objectives throughout the day.

Of course, just because you are hitting your service level and response time objectives doesn't necessarily mean that your customers and senior managers are happy with your center's performance. A center can achieve its objectives, yet still be wasting resources, creating extra work and providing poor quality.

Accessibility is an enabler, not a guarantee of quality or of customer delight. Accessibility means that contacts are getting in and being handled efficiently so that the contact center has the opportunity to accomplish its mission of ensuring long-lasting customer satisfaction and loyalty. If quality is poor, things such as repeat contacts, unnecessary contacts, and escalations and complaints will eventually drive service level down -- and your customers crazy.

Adherence to Schedule

Adherence to schedule is a measurement of how much time during an agent's shift he or she is logged in and handling contacts or at least available to do so. Most centers choose an adherence objective around the 85% to 90% range, meaning that each agent is expected to be available to handle contacts .90 x 60 minutes, or 54 minutes each hour.

Adherence is comprised of time spent in interacting with customers, as well as time spent in after-call work, making necessary outbound calls and waiting for calls to arrive. Time taken for lunch, breaks, training, etc., is not counted as time assigned to handle contacts, and thus is not factored into adherence to schedule measurements.

While always an important metric in contact centers, adherence to schedule has taken on an even more significant role of late as centers have learned to focus more on what really matters, and on what agents can control. Where average handle time and calls per hour used to rule the metric roost, centers have discovered that agents are merely slaves to such measurements, and that these metrics aren't indicative of whether staff is in the right place at the right times, doing what they are supposed to be doing. Agents can't control how many calls are coming in, nor do they have much control over how long frustrated callers' tirades may last, but they can be held accountable for where they are and what they are doing.

This is not to say that contact centers should completely do away with more traditional productivity metrics. After all, the center needs to have an idea about how many calls a typical agent is handling and how long those calls are lasting to help pinpoint any scheduling adjustments that may have to be made. The good news is that when placing a stronger emphasis on adherence to schedule by having agents in the right places at the right times, things like average handle time and calls per hour tend to take care of themselves. That is assuming the center has taken time to provide agents with adequate training and a quality mindset, and that the center has done a decent job of forecasting and scheduling.

That being said, there is the danger of focusing too stringently on adherence to schedule -- watching agents' every move with the help of workforce management technology, and inevitably eliciting cries of micro-management from them.

Here are several recommendations from ICMI for ensuring solid adherence stats that don't cause agents to compare their workplace unfavorably to Orwell's 1984:

• Train each agent on how much of an impact he or she has on the queue, and on customer accessibility and satisfaction.

• Establish concrete service level and response time objectives that everybody knows, understands and accepts.

• Educate agents on the essential steps involved in resource planning to ensure that they understand how schedules are produced.

• Develop appropriate priorities for the wide range of tasks your agents perform on the job.

• Provide real-time information to agents and back it up with training on how to interpret that data.

• Track and manage schedule adherence at the supervisory level, as conditions dictate.

• Track schedule adherence for the entire group for planning purposes and to assess how well management has created a process that enables appropriate schedule adherence.

Forecasted Contact Load vs. Actual Contact Load (a.k.a. Forecasting Accuracy)

This is a performance metric that reflects the percent variance between the number of inbound customer contacts forecasted for a particular time period and the number of said contacts actually received by the center during that time. It is a critical, high-level objective in all contact center environments.

Why is this metric so important? Well, underestimating demand leads to understaffing. This in turn, leads to long wait times in queues, frustrated customers, burned-out agents and high toll-free costs (due not only to the long hold times, but also to the long diatribes by callers livid over the long hold times). On the other hand, overestimating demand results in waste, overstaffing and increased idle time.

Forecasted call load is available from the system used for forecasting (e.g., the center's workforce management system or spreadsheets); while actual call load is tracked by the ACD, workforce management system, email response management system, Web servers -- wherever data is available. Forecasting accuracy should not be reported as a summary of forecasted versus actual contacts across a day, week or month, but rather as an illustration of accuracy for each reporting interval, typically half-hours.

Self-Service Accessibility

It has become the quest of most contact centers to offload as many basic transaction types as possible from live agents to self-service systems -- mainly IVR and interactive Web applications. Doing so helps to not only enhance service efficiencies and cut costs; it also frees up agents to use their valuable skills to assist customers with more complex issues, thus keeping staff engaged and motivated.

But some centers get so caught up trying to lure customers off the phones and into self-service that they forget to track something essential: how well the self-service systems actually "treat" customers.

Self-service accessibility has emerged as a critical metric in this age of automation and customer-centricity. Leading contact centers gauge not only how many customers begin self-service transactions via IVR and the Web, but also how many complete those transactions without live-agent assistance. Centers with a serious focus on effective self-service have invested in tools that record customer interactions with IVR apps and Web sites. Such tools help to pinpoint any system glitches or snags that hinder the customer experience and that cause customers to "zero out" to email or chat with a live agent or simply abandon their online shopping cart. In addition to specialized monitoring tools, some centers have invested in diagnostic technologies that perform proactive load-balancing tests to ensure that IVR systems and Web self-service tools are prepared to effectively and efficiently handle high volumes of contacts -- a sort of self-service check-up.

Many contact centers also survey customers following a self-service transaction to gather direct feedback on their experiences. While not the most proactive or definitive method for gauging self-service accessibility, surveys are a good way to uncover how your valued customers feel about your automated service options.

Contact Quality

This is a very common and critical customer-centric performance metric in all contact centers, regardless of industry, function and size. Top centers track contact quality as a high-level, center-wide metric, as well as an individual agent performance measure.

Contact quality is typically assessed via the monitoring and recording of agent interactions with customers, with quality assurance specialists or supervisors rating the contact using a comprehensive evaluation form that features key criteria that the center feels contributes to a quality interaction from the customer's perspective. Each criterion is usually assigned a numeric value by those conducting monitoring, and weighed based on its impact on customer satisfaction and the center's goals and requirements.

Common quality criteria include such things as:

• Use of appropriate greetings and other call scripts

• Courtesy and professionalism

• Capturing key customer data

• Providing customers with correct and relevant information

• First-contact resolution

• Accuracy in data entry and call coding

• Grammar and spelling in text communication (email and chat)

While monitoring and recording is by far the most common method used for assessing contact quality, some quality measurements may come from ACD-based call coding, from reports generated by customer information systems.

Customer Satisfaction

Customer satisfaction is one of the most critical metrics for any contact center. Studies have revealed, and common sense supports, a critical and direct correlation between customer satisfaction, customer loyalty, corporate revenues and employee morale and performance.

All companies are aware of the importance of securing high customer satisfaction, and all claim to have a strong focus on customer-centric practices, but not all go about measuring customer satisfaction in the most precise and most consistent ways, nor do all have an effective process in place for analyzing and acting on the findings.

While there is no standard method for calculating customer satisfaction, there are certain common practices and processes that enable leading centers to not only effectively and efficiently keep tabs on just how much customers dislike them, but also to make key improvements before customers take their business elsewhere.

Customer satisfaction measurement has evolved beyond clunky mail surveys and phone interviews days after the customer's interaction transpired. The big trend now is to survey callers immediately after the interaction occurs, when the experience is fresh in the customer's mind, and before problems can escalate. Top centers typically do this via IVR-based post-call surveys, similar to the type described in the previous section on first-contact resolution. Callers are asked a series of questions about their interaction with the agent, their feelings about the organization, and their plans to continue doing business with the company. They are asked to rate each question on a numeric scale (often 1 to 5) for easy customer satisfaction calculation. Many surveys also feature a couple of open-ended questions for more detailed customer feedback.

Today's advanced IVR survey apps can be programmed to recognize when a customer gives an abnormally low overall rating and to send an alert to the center manager or quality assurance team. The system can capture (via CTI) the caller's identity, and link it to the actual recording of the call in question for complete analysis of the interaction. After reviewing the survey responses and the call, the manager can quickly call the customer to "repair damage" and hopefully restore trust and loyalty. [For more information on IVR-based post-call surveys, see Greg's "Capture the customer experience by tapping IVR" in our February 2007 issue.]

Of course, not all customers contact the center via phone, thus IVR-based surveys alone are insufficient for holistic customer satisfaction measurement. Progressive centers also gauge how satisfied customers are who have chosen to interact with the company via email or chat. To do so, they send a survey like the IVR-based one to these customers via email, or program the survey to pop up on the customer's screen upon completion of an online interaction.

Achieving Success through Metrics

The metrics your center embraces have a huge impact on the customer experience. And while not EVERY metric can be 100 percent customer-centric -- there are operations costs and business needs that must also be considered -- focusing strictly on straight productivity metrics and managing a contact center primarily as a cost center simply won't wash these days.

Placing a heavy emphasis on all the metrics described in this article will ensure that your company keeps customer satisfaction and loyalty high while simultaneously enabling your center to be an efficient, high-performing, critical business entity; one where agents spend much more time smiling and delighting customers than they do appeasing irate callers.

Service Level Calculations

Today's contact center technology calculates service level in a variety of ways, and it's important to fully understand which calculation your system uses. The following are the most common formulas used for calculating service level:

1. (Calls answered in Y seconds + calls abandoned in Y seconds (total calls answered + total calls abandoned). We at ICMI prefer this calculation because it includes all of the traffic received by the ACD. Calls that abandon before the objective positively affect service level. The calculation provides a fairly complete picture of what is happening to each caller.

2. Calls answered in Y seconds (total calls answered + total calls abandoned). This alternative considers only answered calls and, therefore, is inadequate. The concept of abandonment is entirely ignored.

3. Calls answered in Y seconds (total calls answered + total calls abandoned). This alternative tends to be the least popular among contact center professionals because all calls that abandon negatively affect service level, even those that abandon before the objective. However, some experts point out that this calculation may be appropriate in situations where callers enter a queue after they hear a delay announcement, but not in situations where the caller enters a queue before they hear said announcement.

4. Calls answered in Y seconds (total calls answered + calls abandoned after Y seconds). In this calculation, abandoned calls only impact service level negatively if they abandon after the Y seconds specified. Calls that abandon before the objective do not affect service level. As a result, this is a way to avoid getting "penalized" by callers who abandon quickly, without ignoring abandoned calls altogether.

Read my lips: There is NO industry standard for service level or response time.

The ideal objectives for such metrics vary for each individual contact center, as those objectives are dependent on numerous factors that vary from center to center. The optimum objective is affected by such things as the value of the call, labor costs, trunk costs, customer tolerance and the organization's business objectives. All of these things differ among most contact centers.

The appropriate question managers should be asking isn't, "What's the ideal service level and response time objective for a contact center?" but rather "What's the ideal service level and response time objective for my contact center?"

The appropriate service level and response time objective for your center is one that:

• meets your customers' needs and expectations

• keeps abandonment at acceptable levels

• minimizes agent burnout and errors

• minimizes expenses

• maximizes revenue

• and is supported by senior management.

---

Why Focus on First-Call Resolution?

By Rob McDougall, president of Upstream Works

For most call centers, nearly one-third of inbound calls are repeat callers who weren't satisfied the first time, so it's clear that better FCR (First-Call Resolution) will reduce your costs and improve the customer experience. A focus on FCR will also reduce customer churn and improve agent morale. In that spirit, Upstream Works, a call automation software provider, offers perspective on how you can improve your contact center's performance in getting it right the first time.

First, it helps to understand why FCR doesn't happen:

1. Agents frequently don't have the authority to resolve a problem, even when the solution is obvious. This results in transfers and escalations, with increased hold time and abandons. It also means the next callback will be an escalation call that ties up supervisor time.

2. The agent may not have sufficient training or ability to effectively deal with the customer call.

3. Agents need to find information more easily to provide answers or actions for customers. When the information is difficult to access or unavailable, sometimes agents will guess at the answer or fail to provide an answer, both of which can lead to a callback.

4. The back-end systems might not be up to the task. If the agent makes an address change, but it doesn't propagate through the systems, then the customer will call back.

5. You need clues into customer perceptions and behaviors and why the repeat calls are happening in the first place. This can help, for instance, when you discover customers are calling back trying to get a different result if their account is being suspended.

Here's how to get started:

Step 1. Conduct an information-seeking drill-down with experts who are proven resources. This can help determine the best approach in tackling poor FCR rates head-on.

Step 2. Track repeat calls by analyzing agent logs and the IVR system. Identify whether an issue was resolved on the first call and if not, the actual reasons for repeated calls. Look at all customer interface points, because if the customer first contacted you via self-service systems, that counts as a call.

Step 3. Make certain you provide a means to measure FCR from the first day forward. You need useful information to conduct a root cause analysis. What's one ideal outcome? Could it be one where you continue to use the technology now in place - but making it work better for you?

Step 4. Examine IVR transactions and track where calls go. Look at calls and how the IVR is handling them. Many customers merely opt to zero out. That choice, however, isn't giving you interpretable data. Callers choose to zero out for numerous reasons, including their desire to speak with a live agent.

One method of measurement can be a standard question agents can ask the customer at the close of a call:

• "Is there anything else I can do?"

• "Is this what you wanted?"

These questions allow a crude estimate of FCR. Ideally, your contact system allows the answer to be documented and summarized for the center.

---

Copyright 2007 CMP Media LLC. All rights reserved. 3/1/07, Issue # 2003, page 24.

---