Building Customer-Centric Metrics

Aligning Operational Excellence with Customer Experience

Joe was a vice president at a large tech corporation who was responsible for managing over 40 call centers that handled over 200 million calls per year. Scribbled across a white board behind Joe’s desk was the current status of over 15 different call center metrics. Pointing to the white board I asked Joe, "Which one or two of those metrics are most important to you?" One by one, Joe explained the importance and equality each metric had in managing his call center operations.

To Joe, no metric was more important than any other. And just like Joe, many business areas regularly report their metrics to help manage their operations. But how do we know if we have the right measurements? How do we know our measurements are reliable or accurate? How do we know which metrics are most important? The key is designing metrics that are centered around the customer (i.e., customer-centric metrics). Let's explore that further and apply it to a call center operation like Joe's.

All operational metrics should begin with the customer

To begin with the customer, we need to understand what the customer wants and expects. Whether someone is buying a product or a service, every customer in any situation always asks the same thing: "Am I getting the right THING at the right TIME for the right PRICE?" This single question represents the voice of the customer or VOC and contains three key perspectives:

• Right THING = "Am I getting the level of quality or accuracy I expect?"
• Right TIME = "Am I getting it when I want it?"
• Right PRICE = "Am I getting value for the right amount of my time and money I’m investing?"

Although every customer asks this question, the expectations will vary between people and even for the same person. For example, we can buy a cheeseburger at a fast food restaurant or a high-end steakhouse restaurant. Although the ingredients are generally the same, the final product and the expectations of the customer may vary greatly. At a high-end restaurant, a customer usually expects to wait a little longer and pay a higher price to get a burger that is higher quality, but that same person at a fast-food restaurant may be willing to compromise the quality to get the burger faster and cheaper.

How do we ensure our business is meeting our customer's expectations?

I believe every business should be asking a single question that's similar to the customer's VOC question. That is, "Am I doing the right THING at the right TIME in the right WAY?" This perspective reflects the voice of the process (VOP) to let us know how capable our process is to meet those customer expectations. Just like the VOC question, this VOP also has three key perspectives:

• Right THING = "Am I delivering to my customer the quality and accuracy they expect?"
• Right TIME = "Am I delivering results to my customer when they want it?"
• Right WAY = "Am I delivering that right THING and TIME with the most efficient process?"

As you can see, the VOC and VOP questions are very similar, but a key difference is between customer investment (right PRICE) vs. process efficiency (right WAY). These are indirectly related because if a business delivers the right THING at the right TIME with an inefficient process, then that higher cost of inefficiency may influence a higher price to the customer (either by charging a higher rate, or requiring more effort from the customer, or in lowering the profit margin). However, if a business can deliver that right THING at the right TIME with a very efficient process, then they may use the cost savings in a variety of ways like reducing the price to build their customer base, or improving their quality, or enhancing their product/service features, etc.

Balancing how we measure the right THING, TIME, and WAY

This VOP perspective of delivering the right THING at the right TIME in the right WAY should be the basis for defining customer-centric metrics. However, not all 3 should be treated equally. Delivering the quality and timeliness the customer expects is more important than operational efficiency. This may seem counter-intuitive, but let me explain why.

First, we must succeed on all three VOP perspectives of delivering the right THING, TIME, and WAY. Failing on any of them creates a risk of failing to meet the customer's expectations and ultimately losing that customer. The chart below illustrates this:

In the examples above, failure in any of the three VOP perspectives can be disastrous. However, failure to do the right THING and TIME are most important because they directly affect the customer's experience. A customer may not know or even care how inefficient our process is as long as we deliver the right THING at the right TIME. Therefore, the way we measure these two perspectives should be of primary importance.

That doesn't mean we should ignore efficiency. It only means that the way we measure efficiency should be of secondary importance and our level of process efficiency should never supplant the higher priority of our meeting the customer's expectations for the right THING and TIME.

How do we define the ideal customer-centric metrics?

Below is a 3 step approach that can be used to find the ideal customer-centric metrics that reflect the VOP perspective.

Step 1 - Select one of your types of customers - First think about the type of person who most often buys your products or services, such as their income level, where they live, their age, race, gender, education-level, etc. This is your customer segment. It doesn’t mean your products or services exclude other types of people. The point here is to think of the most typical kind of customer you serve overall, then later you can re-run this approach using a different kind of customer.

Step 2 - Identify your key output(s) given to these customers - For the customer segment you defined in step 1, think about what they are buying from you. Is it a certain type of product or service? Are there any other products or services you offer that may be associated with that same customer segment?

Step 3 - Fill out the customer-centric matrix - Fill out the matrix below to guide you to what that customer segment expects for the product or service you’re selling. Start by thinking of the level of quality or accuracy that customer segment wants for that particular type of product or service. Then ask yourself the 4 questions listed across the top.

Applying this 3 step model to a call center operations example

Ask any seasoned call center expert and they'll likely know the typical metrics used for measuring call center operations such as average handle time (AHT), average speed to answer (ASA), issue resolution (IR), customer satisfaction (CSAT), net promoter score (NPS), etc. For this example, we'll skip steps 1 and 2 by presuming this is for a generic customer and output, and we'll jump to step 3 to complete the customer-centric matrix as shown below.

Sometimes the most challenging part of this approach is identifying the gaps/risks for the current metrics. Those metrics may be so embedded into the operations that it can be difficult for some people to challenge them.

In our scenario, there are inherent problems with some of the key call center metrics like those relying on survey-based data (CSAT, IR, NPS, etc.) and on time-based data (like AHT, ASA, etc.). For example, surveys are often not aligned to the population by key customer characteristics (e.g., gender, location, product/service purchased, tenure, etc.). As such, a minority group of customers may be influencing results for the majority. Also, I've observed where customers are wrong when surveyed whether or not this was their first time calling, or they score the survey incorrectly by misinterpreting what’s good vs. bad on a numerical scale. And it's not just the customers, phone agents who are held accountable to certain metrics may manipulate the results (e.g., reduce AHT by transferring or shortening a call).

Does this mean we abandon these metrics? No, but we should see if there's a better way to measure without the same risks. For example, we could tie the quality measurement to a customer's behavior like repeat calls or purchasing habits. If a customer's issue was not resolved on a call and their issue persists, then they are likely to call back. We can find these by measuring repeat call activity by customer. Or we can measure a customer's behavior after the call by seeing if they made additional purchases in the same pattern as their prior purchases. 

We may be our own worst enemy

Sometimes our current measurements fight against us. For example, a customer who calls two times at 500 seconds per call yields an AHT of 500 seconds. However, if we measured the repeat call activity, we'd see that customer experienced 1,000 seconds on the phone with us. Instead, why not allow more time on the first call to ensure the issue is resolved? If a customer makes one call for 800 seconds and doesn't call back, the AHT will look much worse at 800 seconds. But from the customer perspective, it's far better because they experienced 200 fewer seconds to get their issue resolved. In this way, a measurement like total handle time (THT) can be used to understand the total affect by customer across all calls they made.

A key part of improving call center operations is to understand the call reason, or the customer's issue that caused them to contact us. This is often done using self-selecting options by phone agents or by evaluating agent notes on a call. However there are many newer solutions like voice analytics that are maturing in accuracy and reliability to detect call reasons. These tools often include other valuable insights such as customer sentiment and agent behavior.

How do we know if we identified good customer-centric metrics?

Good and bad metrics will tend to have some unique characteristics. Below is an example of these characteristics:

So what about Joe and all his other metrics?

Joe had moved on from his role because it became so difficult to effectively lead the call center operations with so many metrics. Joe was treating call center operations like an airplane cockpit where there are so many measurements that it required everyone to be an expert in how to interpret them.

Instead, the few customer-centric metrics we build should be like a car dashboard where they’re simple, intuitive, highly visible and treated as a top priority for driving the operations. Like a check engine light, all other metrics should be designed to trigger only when critical issues arise. Otherwise, we need to keep these secondary metrics in the background so they don’t distract us from the primary metrics that matter most in delivering value to our customer.

About Matt and StatStuff

Matt is a Lean Six Sigma (LSS) Master Black Belt (MBB) who's been leading continuous improvement teams for over 20 years across several industries and across many departments within several large corporations. Matt developed and led several enterprise-wide LSS training and certification programs and founded StatStuff as the only FREE source for complete LSS training. StatStuff is highly endorsed as quality LSS training from leaders at top companies like Apple, eBay, Pepsico, Bank of America, Dell, Sprint, BP, Honeywell, etc. Many companies, training organizations, and universities use StatStuff for their training curriculum.