How to Track Customer Service With Your Analytics

Trying to make smart decisions about your business without metrics is like driving your car with your eyes closed.

While this thinking seems to permeate most marketing departments, it also is incredibly important to view the quality of your customer service through this lens as well.

Bad customer service can be a huge leak in your bottom line, and it’s a big problem that most companies have. The fact that only 4% of unsatisfied customers ever speak up means that a lack of customer complaints doesn’t necessarily prove you are providing outstanding service.

So how can you get started tracking the effectiveness of your customer support? Is it even worth the effort?

Today, you’ll find out how to answer both questions using the new KISSmetrics integration with Help Scout, the invisible help desk software.

Here’s what we’ll be going over:

  1. Which customer service metrics actually matter
  2. Implementing data-driven support with Help Scout + KISSmetrics

So strap yourself in, and let’s get started!

The Customer Service Metrics that Matter

There are vanity metrics in nearly every aspect of your business, and customer support is no different.

Remember that it’s people who pay you, not the superfluous stats. Since that’s the case, it’s important that you focus on the customer service metrics that will have the biggest impact on your bottom line.

Here are the important metrics you should be keeping an eye on:

1. Total Volume by Channel

The total volume of all of the support queries you receive is incredibly important to track. As chief happiness officer for Buffer, Carolyn Kopprasch, describes this importance by stating:

The moment to start hiring is before you start scrambling.

Having a close tally of your total volume will be important for many other metrics, too, since it serves as a baseline figure you will be comparing against.

One thing to note here is that it is to your advantage to pick the “channel” that works best for your business. Our friends at Wistia unabashedly removed their phone number from their site’s homepage because they found that solving tickets as a team via email actually allowed them to offer a better experience for their customers.

wistia removed their phone number

You have to work with the channel that works for you and your customers, and which will allow you to scale as your company grows.

2. Response Time

While speed is not what customers care about most, it still is very important. Slow service is viewed by customers as uncaring and incompetent.

One important metric to track within this data is your 24 hour response rate, or your ability to answer a ticket within 24 hours. Let’s be honest, when it comes to email support, customers will (rightfully) not want to wait more than a few hours, and certainly no more than a day, to hear back from you.

While a few consumer studies have noted that customers do have quite a bit of patience – 50 percent of consumers give a brand one week to respond to a question before they stop doing business with them – providing “okay” support really isn’t okay, especially if you want to stand out.

In fact, to give you an idea of how far service quality has come and what the best companies are working to provide these days, the Buffer support team is looking at improving their 1 hour response rates.

3. First Contact Resolution Rates

Having the fastest response rates around won’t do you any good if your response quality suffers.

This is especially true for customer complaints, as data from Lee Resources shows that up to 95% of customers will give you a second chance if you handle their initial complaint successfully and in a timely manner.

Tracking and actively improving how many interactions your team resolves on the initial engagement with a customer will go a long way toward improving the overall customer experience your business is able to offer.

4. Hold Time and Abandonment Rates

How many customers are abandoning your live chat or help line?

For chat, especially, response time is crucial, as some recent consumer data shows that 71% of customers on live help expect assistance within 5 minutes, and 40% of those folks want help immediately.

Since most (up to 60%) will outright abandon a site if they don’t get hold of someone, you can lose potential prospects and upset otherwise happy customers if you don’t keep a close eye on these numbers.

Once a customer has contact, you can slow things down and solve their problem with competent and personal service. But starting the conversation quickly is a must.

5. WOW Moments

It’s harder to quantify your team’s ability to WOW customers, but WOW-factors are essential in building a word-of-mouth presence among customers and to keep them coming back.

Zappos has an interesting model known as the Happiness Experience Form that they use to measure 4 critical factors:

  1. Did the agent try twice to make a personal emotional connection (PEC)?
  2. Did they keep the rapport going after the customer responded to their attempt?
  3. Did they address unstated needs?
  4. Did they provide a “wow experience”?

While this isn’t the kind of metric you can evaluate for every interaction, you certainly can pick and choose several when reviewing performance or training new team members. Imagine going over similar questions as you train new employees or interns. It helps both sides optimize for the WOW-factor.

Don’t forget, qualitative data (measuring quality) is just as powerful as quantitative when you measure customer service. The kinds of metrics that don’t fit in a spreadsheet often are the most important.

Data-driven Support with Help Scout + KISSmetrics

Without a doubt, all of these metrics help you measure the company’s customer service efforts. But there’s one critical piece missing, one data point every entrepreneur is trying to uncover: ROI. Why waste the time optimizing all of these metrics unless you feel that over-achieving can build massive value for the business? That’s what being data-driven is all about.

By tying together KISSmetrics with Help Scout, you can get a clear answer to the all-important ROI question.

Your help desk contains most of the metrics we’ve already talked about. KISSmetrics contains account data about your customers: their plan, lifetime value, and all of the lifecycle activity that’s important to your business. Putting these two collections of data together is where the magic happens.

When KISSmetrics receives data from Help Scout, it can be tied to a customer and all of their lifecycle data, enabling you to determine the following:

  • When should you consider hiring another customer support employee?
  • How much does it cost to support the average customer per month?
  • What percentage of support inquiries are from free customers? Paying customers?
  • How does support volume ebb and flow over a customer’s lifetime?
  • Do high-volume customers share any lifecycle or profile characteristics?

km helpscout integration

Who says you can’t be data-driven about customer support? By integrating KISSmetrics with Help Scout, you certainly can be. Most important, you can have unprecedented insights into what role support plays across your customer base.

Using tools like Help Scout and KISSmetrics together will cover your bases on the quantifiable side of support, but, as we’ve already mentioned, the WOW-factors can’t always be counted that way.

Qualitative data is equally important in measuring ROI with regard to the performance of your team and business as a whole. We wrote a complete eBook about it called The Business Case for Loving Customers. It even includes a fascinating case study about how KISSmetrics does customer support.

Over to You

Now you know exactly how to get started applying data-driven insights to your customer experience.

Making smart decisions about how you provide and measure quality service offers a unique opportunity for your business to out-support the competition.

This allows you to build your business around what matters most: loyal customers!

Let us know what you think about this latest KISSmetrics integration by leaving a comment below!

Did you like this article?

Feel free to spread these stats:

  • 50% of consumers give a brand one week to respond to a question before they stop doing business with them. »tweet«
  • 95% of customers will give you a second chance if you handle their initial complaint successfully and in a timely manner. »tweet«
  • 71% of customers on live help expect assistance within 5 minutes, and 40% of those folks want help immediately. »tweet«

About the Author: Gregory Ciotti is the content strategist at Help Scout, the invisible help desk software for startups and small businesses.

  1. I love the Zappos philosophy and their emphasis on wow moments! Great tidbits here, great article! Customer service is essential, and we have no excuse to do it poorly if we want to “wow” customers.

  2. I love your advice on this topic. Thanks for sharing.

  3. I very mach love your tips .I hope next time you provide your fan useful tips.

  4. Great customer support is always a plus point. Even if the product/service is not that much great but that can win people with great customer support :). Thanks for your tips :)

  5. Analytic s is a great method for tracking customer service. Special thanks for the author to provide a praiseworthy tutorial. I like this site very much.

  6. I love the “WOW moment”-metric! Of course its difficulties in measurability make it a little unconventional as a metric, but when hasn’t being a little unconventional helped so far… what a great company philosophy!
    I think you can’t stress the importancy of a great customer experience in today’s eCommerce enough – just one look on your statistic insights makes that clear once more. Thank you especially for those interesting numbers! Will have a little tweet with them straightaway. :)

  7. Hi, Can you recommend any application which can track customer emails response time on emails & other metrics that you have defined above.

9 comments

Please use your real name when commenting. Otherwise, your comment may be deleted.

← Previous ArticleNext Article →
Buffer