Enlightened customer support teams — you know, the kind behind the help desks of most modern saas tools — are more focused on creating a great experience for customers than they are on how cheaply they can manage an outsourced call center. And yet, all too often a support team’s metrics focus myopically on a customer’s interactions only with support. They track things like customer satisfaction (CSAT), first reply time (FRT), and first contact resolution (FCR) and as these numbers improve, they rest easy, assured in the knowledge that customers are having a great experience… with support.

Some support teams are starting to use more holistic metrics like Net Promoter Score (NPS) and the HBR-blessed Customer Effort Score (CES) to better understand how customers feel about their product rather than just their support experience. This is definitely a move in the right direction, and it makes sense for the team that talks to customers all day, every day to report back on the general level of satisfaction with the product. However, even with more customer-experience focused numbers, there’s still a significant amount of insight left on the table. Let’s say your NPS and CES scores come back okay…ish, and you want to make some improvements. Where do you start? We’ve been thinking about this a lot at FullStory and we have some ideas.

Better metrics for your arsenal.

Ultimately, while we do want to keep an eye on how well our support processes are working, we really want metrics that reflect the customer’s actual experience with our product. NPS and CES are a great place to start, but we’ve found that tracking more of the things that serve as inputs to those numbers gives us a clearer picture of** why** our scores are what they are and what we can do to improve them.

If you’re interested in doing the same, here are some ways to get started:

1. Number of tickets per account.

How frequently are customers having to contact you?
We like this metric so much at FullStory that we report on it weekly at our all-hands meeting. Tracking this number gives you a great sense of how easy (aka frictionless) it is to use your product. If your NPS and CES scores are lower than you’d like and your customers are writing in multiple times per day/week/month, spend a cycle or two focused on reducing this metric and watch how those Big Numbers(™) change.

How can you measure it?
To get a quick sense of where you stand, simply take the number of tickets you received for a given time period (we measure this weekly) and divide it by the number of active accounts or customers you have. (One bonus of this metric is that, as opposed to the rate of tickets per day or per week, the number is normalized even as your customer base grows larger and larger.)

To take a more insightful (but decidedly un-bionic approach, get your spreadsheet skills ready:

  1. Export your ticket IDs along with the customer’s name (or the account name) for a set time period.
  2. Create a pivot table with a count of the number of ticket IDs to each unique name.
  3. Average the number of tickets per customer or account.

Bonus: Since you’ve done the extra work, you can also see which customers have had to write into support much more often than normal. These guys might deserve a little extra attention.

How can you improve it?
Here’s the fun part. Once you’re ready to make some headway, you’ll want to dig in to understand what’s driving the number of tickets. If your team uses a tagging system (and is diligent about it), you can quickly get a sense for the key drivers.

Are you seeing lots of questions early in a customer’s life cycle? Take a closer look at your onboarding program. Schedule some time to Game Film people’s first few sessions with your product. Within just a few minutes you’ll start to see areas that consistently trip up your new customers.

Are customers writing in with questions that are “clearly” answered in your knowledge base? Investigate ways to make that content easier to access. Watch their session to confirm whether the problem is finding the article or the article itself. Better yet, make the task itself more self-explanatory and reduce the need for that help article all together.

Dealing with a lot of bug reports? Time to enlist some engineering time toward fixing those persistent product issues that just won’t go away. Which brings me to the next metric…

2. Number of support tickets per bug

How small are those issues really?
As a customer, when you run into bugs in a product you’re using (even if they’re small bugs that don’t stop you from accomplishing your goal) your feeling about the brand is affected. No one wants a buggy product, and yet prioritizing bug fixes can be difficult. Even when you can devote time, how do you know which bugs to fix first? Having some hard numbers on which bugs are driving the most tickets, while not a perfect approach, at least provides some additional context to use in the prioritization process.

How can you measure it?
Applying tags consistently and being diligent with whatever bug tracking tool you’re using (Bugsnag, JIRA, Aha!, etc.) is a great place to start. It requires a good amount of cross-team coordination — deciding on which tags to use and then, you know, using them correctly and whatnot–but the extra effort is worth it. Now you can say with certainty that an issue consistently deemed “small” is the driver of 40 (or 400) support tickets every month.

How can you improve it?
FIX BUGS, DUH. I know, I know, easier said than done. We too struggle with bug squashing when there are so many shiny new features dying to be built. But we’ve taken a new approach recently that is working well so far. Each cycle, we take one or two engineers and have them focus exclusively on fixing the bugs that drive support tickets. For the entire seven weeks (our cycle length), they partner directly with support and take on no additional engineering tasks. Since we’ve been doing this we’ve noticed a big difference in the speed with which we can address customer-facing bugs.

3. Tickets resolved with no additional follow up required.

We track first contact resolution (FCR) internally and we really like this metric. It’s very rewarding to read a customer’s ticket, watch their session, understand exactly what’s going on, and provide a clear answer to their problem all in a single response. But sometimes FCR can be misleading. A ticket solved in one response doesn’t necessarily mean that the issue itself is resolved. The customer might hit the same issue again a week later, generating a new support ticket (rather than a trackable follow up to the same ticket). When we look at our FCR metric we think we’re doing great, but the reality is that our customers aren’t having a great experience.

What we really want to understand with FCR is how many tickets are resolved quickly and require no additional follow up.

How can you measure it?
I’ll be honest with you, this one is tricky. You can use tags to indicate when a customer has written in with the same issue in the past. Do a quick check of their support history and add a tag like “repeat issue” or link the new ticket to their previous one. Ideal? No. Insightful? Certainly.

If you haven’t been tagging things already and you’re anxious to get a glimpse into how often customers write in with the same issue, try doing a quick search for the word “again” in your support tool. You’ll get a few false positives from customers saying things like “Thanks again!” or, if you’re lucky, “You guys have totally wowed us ONCE AGAIN.” But you’ll also see some frustrated customers struggling with different aspects of your product.

How can you improve it?
Improving the drivers of customers’ repeat issues depends on what those pesky problems actually are. But whether it’s a confusing part of your UI or a persistent bug, understanding how big of a problem it is and framing it well for the rest of your team or company can be a big help. If data on the number of tickets and repeat tickets still isn’t enough, try sending them playbacks of customers hitting the issue repeatedly. It’s really true what we say about being face-to-face with the customer’s experience in your product: it dramatically raises your level of empathy and, as a result, your motivation to solve the problem.

Support your customers, not your numbers.

Regardless of the metrics we choose as our guideposts, what we really care about is the customer experience. Not just because we’re empathetic people and want our customers to be happy (although we are and we do), but because we know, like the hospitality industry and brick-and-mortar retailers have known before us, that happy customers are loyal (read: pay more $$$), are more likely to recommend (read: bring in new customers), and, ultimately, the best products with the best experience will win more customers’ hearts.

We certainly don’t claim to be experts at customer support. We take a different approach that we’ve talked about before, and we think a lot about how we keep tabs on our customer’s actual experience (we built FullStory to help ourselves and our customers do exactly that after all).

How does your support team approach these types of metrics? We’d love to learn what you’re tracking and why. Let us know in the comments or on social media.