We’re on a constant treadmill of measurement, augmentation, and optimization. And much like stepping onto the scale, KPIs tell us what’s working and what’s not. For many businesses, KPIs rule supreme. For customer support, KPIs are the heart monitor for the organization. But many organizations that focus solely on heart rate believe themselves to be healthy while high blood pressure remains unchecked. As Arthur Conan Doyle said: “There is nothing more deceptive than an obvious fact.”

Your metrics aren’t altogether wrong, but they aren’t pointing you in the right direction either…

So is everything you know a lie? Maybe not. Rather than looking at them as lies, focus on these top 3 metrics that actually matter, and learn how to make them work for you:

Support Ticket KPIs You’re Probably Getting Wrong

These ticket based KPIs typically point to a support center’s efficiency. These are important metrics to track, but they might not indicate what you think they do.

For example:

Overall Ticket Count and Tickets Resolved

This metric usually indicates Incoming ticket volume. Ticket count is a key factor for determining ticket resolution rates. However, it should also account for:

- Effectiveness of UI/UX.

- High similar ticket counts: multiple tickets on similar issues can point to problems in your self-service areas.

- Low ticket counts and high churn: this indicates that your customers don’t know how to access your support systems.

Average Ticket Resolution Time

This metric usually indicates the efficiency of support staff, and their ability to meet support levels. However, it should also account for ticket difficulty, as well as the customers' ability to resolve their issue through self-support methods.

Metrics that Matter: Customer KPIs

If you’re meticulous about customer-centric KPIs, that’s great! Your customers are your reason for doing business. See what they can show you.

• Channel Distribution: Your customers should dictate your support hours. If your live chat numbers are low compared to phone inquiries, then emphasize live chat as a simple alternative to boost your team’s efficiency. Conversely, be sure that you have a sufficient team size as determined by customer volume per channel.

• Customer Satisfaction: Some consider a low churn rate or absence of direct complaints to be the hallmark of satisfied customers. Others email surveys that say, “How did we do?” But it’s clear that the company has no interest in making actionable change.

To determine satisfaction, look at your social media. It’s not an “objective” KPI per se, but it’s a good indicator that your customers are happy if they’re saying good things about you. Do they share your blog posts? Do they rave about your service? But more importantly, do you interact with both positive and negative social media attention? Positive brand sentiment on social media platforms should be encouraged through dialogue and consistent social media attention.

Top Customers in Need

You should be able to name your top 5 or 10 customers. But what about those customers who require a little more attention? Your customers who require more attention should also require management’s attention. They might not have created the most tickets, but they cause headaches. They request highly technical solutions that cause long wait times, which can create a bottleneck for your support team. They respond to surveys with low satisfaction. They submit a high volume of finicky tickets.

Look at the customers that you’ve lost in the past year or two. Do these customers have anything in common? Consider whether it’s wise to exert effort to fix your processes that cause churn. You may simply recognize that these “customers in need” might end up costing more time and money than is responsible to exert. These customers are highly likely to churn– it’s up to you to recognize which metrics indicate churn and act appropriately.

It’s easy to fall into a KPI rut. It’s easier to get tunnel vision with your metrics. Revisit your KPIs. Do they say what you think they do? You might be surprised.‍