A few weeks ago I wrote about the importance of effective customer service. But how do you know if you are delivering that? And what channels — phone, email, chat, social media, etc. — should you emphasize? How about innovations like chatbots, intelligent agents, and other automated systems?
Measuring customer service breaks down into two general areas: satisfaction metrics based on customer attitudes and performance metrics based on support staff behavior.
The CSAT Metric
The primary way that companies gauge customer service satisfaction is by asking customers for their opinion on a scale from 0 to 10. This customer satisfaction metric (often referred to as “CSAT”) averages out all scores and presents them as a percentage. For a pinpoint example, in a Forrester report, available here from Statista, CSAT scores for streaming video services ranged from 58% for Netflix to 44% for HBO Now. For a very broad example, consider this graphic showing satisfaction with call center service, that is, customer service by phone, across the entire globe:
(Source: CFI via Statista)
The same source, CFI Group’s Contact Center Satisfaction Index, breaks down its data by industry:
(Source: CFI via Statista)
If you want more specific benchmarking data for your own market and industry, you may have to seek it from a research firm or trade association. And of course, tracking your own metrics over time — and hopefully showing improvements — is the primary positive way to use this kind of data.
One caveat regarding CSAT use: Turns out there are cultural differences in the ways that users react to 10-point scales. For example, American customers are more likely to award 9s and 10s than Japanese customers for the exact same service. Here’s how Tamina Steil at Userlike explains it:
- Your CSAT score is then the average rating of your customer responses. The scale typically ranges between 1 – 3, 1 – 5, or 1 – 10. A larger range is not always better, due to cultural differences in how people rate their satisfaction. An article in Psychological Science, for example, showed that people in individualistic countries choose the more extreme sides more frequently than those in collectivistic countries. … Simpler scales are more robust to cultural differences and more suited for capturing service quality.
So if your customer base is global, consider measuring customer service satisfaction on a scale of 1 to 5, or even 1 to 4. For example, check out this four-choice measurement system from feedback.usa.gov:
Net Promoter Score
Another widely used customer satisfaction metric is net promoter score. This is based on CSAT data, comparing the extremes and ignoring the middling results. Pascal van Opzeeland, also writing at Userlike, offers more detail:
- NPS divides your customers into three categories: Customers answering with a score of 6 or lower are segmented as “Detractors”. They won’t recommend you to anyone, will probably not buy from you again, and might even hurt you through negative word-of-mouth. Those with a 7 or 8 are segmented as “Passives”. They are quite satisfied, but not ecstatic enough to recommend you. Those with a 9 or 10 fall into the “Promoters” segment. They are your groupies, your equivalent of the people camping in front of the Apple store. They’re likely to recommend you and buy from you again. Your total Net Promoter Score is calculated by subtracting your “Detractors” percentage from your “Promoters” percentage.
As an example, look at the wide range of NPS results in a survey of U.S. lenders:
(Source: World Economic Forum via Statista)
While the net promoter score is a powerful metric, there are a few caveats. An emphasis on this metric can push customer service workers to strive for 9s and 10s, aiming to truly win over customers, which sounds great, but which research has shown may not offer the best ROI. According to the Harvard Business Review, the cost to push customers to 9s and 10s is higher than it is worth because those ranking service so high are no more likely than 7s and 8s to remain customers. Referencing research by the Customer Contact Council, Matthew Dixon, Karen Freeman, and Nicholas Toman at the Harvard Business Review article declare:
- Delighting customers doesn’t build loyalty; reducing their effort-the work they must do to get their problem solved-does. … According to conventional wisdom, customers are more loyal to firms that go above and beyond. But our research shows that exceeding their expectations during service interactions (for example, by offering a refund, a free product, or a free service such as expedited shipping) makes customers only marginally more loyal than simply meeting their needs.
Bottom line: Do not neglect basic prompt complete customer service in order to recruit evangelist customers.
Customer Service KPIs
Beyond CAST and NPS, there are other notable customer-centric satisfaction measures that can evaluate the effectiveness of customer support. First contact resolution measures the frequency with which a customer issue is resolved the first time the customer calls. The abandoned call rate tells how long customers wait before they hang up because they are on hold. Certainly, customers hate to be on hold. Consider this research on how long they are willing to wait:
(Source: American Express via Statista)
Of course, abandoned call rates are easier to quantify than customer opinion. And so are a large number of ways to track customer service work by employees. Here are a selection of metrics. For further details, click on each one.
- Cost per Contact – Operational costs on a per call basis.
- SERVQUAL – Periodic measurement of reliability, assurance, tangibles, empathy and responsiveness.
- Occupancy (or Utilization) – Percent of time staffers work on actual customer contact.
- Average Number of Replies Per Case – Measures team effectiveness.
- Average Response Time – The time it takes to satisfy the customer. Look at this graphic showing data from 1,000 companies:
But what’s the best way for customers to reach you? Although customers are more and more willing to try alternatives, the venerable telephone remains the medium of choice. When asked for an opinion, customers overwhelmingly say that phones offer the fastest response time:
(Source: The Northridge Group via Statista)
However, although they think phones are the fastest, customers also think that companies are pushing them to other media:
Customers are more likely to say that companies make it easy to use chat and email than they do to use phone support. People understand that good phone support is more costly to supply than chat and email. When a company makes its phone number hard to find, but offers easy access to other media, the company may look cheap, and it takes the risk of decreasing customer satisfaction before the customer has even reached out to make an initial contact.
But cutting costs may be the most important priority, or the difficulty of scaling support services may also be a huge challenge. With that in mind, it is no surprise that chatbots, intelligent agents, and fully automated systems are on the rise.
Looking at both advantages and hurdles, the State of Chatbots Report 2018 asked poll respondents about the benefits they would hope to enjoy by using chatbots:
0)] (Source: Drift; SurveyMonkey; Salesforce.com; myclever via Statista)
The same research asked customers about the reasons they would resist chatbot support:
1)] (Source: Drift; SurveyMonkey; Salesforce.com; myclever via Statista)
With so many customers simply preferring human contact, how effective can chatbots actually be? Here’s some data on that:
(Source: Microsoft via Statista)
The answer is: Meh. People are mostly on the fence about their chatbot experiences. Still, to a fellow who fondly recalls the golden age of MovieFone, seeing what Alexa and Siri can do now, there’s no ruling out a future in which customers can’t tell the computers on support duty from the humans.
Don’t just pay lip service to the need for good customer service. Rather, measure the effectiveness of your support services, and then use your benchmark data to do even better. Moreover, consider the utility of different ways to reach out to your customers. That can even include automated systems, especially as they prove themselves in the years ahead.