When Do We Break the Rules−and When Do We Follow Them?
In his Psychology Today article “When Breaking the Rules is the Smart Thing to Do,” author and change leadership consultant Gustavo Razzetti says while most people are afraid to break the rules, others think rules are meant to be broken. Undoubtedly there are rule-followers and rule-breakers, but I suspect many of us are somewhere in between. Take the speed limit, for example. Do you follow it, ignore it, or create your own 5-miles-over-the-limit rule?
In my previous blog, I wrote about breaking the rule of using traditional metrics to manage your contact center. For instance, managing to Average Speed of Answer is not as critical as in the past, provided you give your customers alternatives such as the ability to request a callback.
Why Mastery Is Essential
So when do we break the rules, and when do we follow them? Razzetti says “Master it before you break it.” Sage advice indeed. Razzetti gives the example of Pablo Picasso, the famous Spanish painter who created the cubism art form. Picasso first mastered traditional drawing and painting; the Musee d’Orsay in Paris shows the progression, and his first paintings are dramatically different from his later work.
Translating the “master it before you break it” approach to the contact center, this means we first need to fully understand each metric we use today to measure performance. What purpose did each serve when it was first established as a measure of success? Are there emerging technologies that change the importance or the impact of that metric? What additional data should come into play?
Let’s look at a couple of examples.
Minding Your Contact Center Metrics
Perhaps you manage to the Average Handle Time (AHT) metric or Average Talk Time, and encourage agents to keep their handle times low. As discussed in the previous blog, on-line retailer Zappos decided the AHT metric did not support their business model given Zappos’ focus on optimizing every single customer experience. But if your business is different, and it makes sense for you to keep AHT below a certain level, you may want to consider the impact of Average Wrap-up Time (AWT). That’s the true picture of time an agent spends supporting a customer. You could have an agent who has low handle times, but spends far more time than his peers on wrap-up work. So, rather than managing to AHT, you may want to create a custom metric (AHT+AWT) that enables a more accurate assessment of how much time an agent spends supporting each customer.
As suggested above, as you evaluate traditional metrics you’ll also want to assess what additional data should be taken into consideration along with that metric. For example, let’s look at AHT through a different lens. Does your company have a focus on customer satisfaction, or improving the customer experience? If so, offer a post-call survey to your customers and look for a correlation between AHT and survey results. Perhaps an agent who has a longer-than-average handle time also has extraordinarily good rankings from the customers he has served.
Correlation and Causation
A correlation between variables, however, does not automatically mean that the change in one variable is the cause of the change in the values of the other variable. Be sure to assess for causation to learn whether one event is the result of the occurrence of the other event. Why? Take a look at this spurious correlations web page and you’ll understand. For example, there is a strong correlation between per capita cheese consumption and the number of people who died by becoming tangled in their bedsheets. And there’s an even stronger correlation between total revenue generated by arcades and the number of people getting computer science doctorates in the U.S. Strong correlations in both of these instances? Yes. Causation? Extremely unlikely.
Going back to our earlier example regarding the correlation between AHT and survey results, you could drill down to details (perhaps listen to some of that agent’s longer calls) to learn whether his longer handle times did indeed cause higher satisfaction ratings (causation). If so, it wouldn’t make sense to penalize him for long handle times. Instead, you’d likely want to learn more about what he is doing to make customers happy, and train other agents in his techniques.
Back to our initial question - when do we break the rules for measuring contact center success, and when do we follow them? Alas, there is no one-size-fits-all guidebook.
But there are some guidelines you may want to follow:
- Identify your company’s business goals that pertain to customer experience
- Assess what your contact center needs to do to support these goals
- Ask whether the metrics you are using today enable you to measure success
- Analyze each metric not in isolation, but in combination with other relevant metrics
Really, the bottom line is this: metrics are subject to change and we should be open-minded. If a metric no longer serves us, then we should consider giving it up or looking at it in conjunction with other measures.
At a fundamental level, evaluate whether your current contact center solution enables you to easily customize metrics and quickly build the custom reports that enable you to fully and intelligently assess contact center and business success. By first mastering the traditional contact center metrics and then 'breaking' them to align them with your business, you may find that breaking the rules means more customers and higher revenues.
Let us know how we can help!