It’s easy to fall in love with data. Data, metrics, and analytics settle arguments; they inform decisions, predict the future, and add relevancy. It’s why companies rely on metrics such as CSAT, CES, and NPS to assess the quality of the customer experience and identify problem areas. The catch: Even data may not tell the whole story. Using three popular metrics, we’ll explain how to avoid mistaking the part for the whole when it comes to measuring the customer experience and how to deliver even stronger results.
1. Customer effort: got context?
It’s not a mystery why companies want to eliminate high-effort customer interactions, such as transfers, requests to repeat information, and long wait times. These are poor experiences that only add to a customer’s frustrations. What’s more, 96 percent of customers with a high-effort service interaction become more disloyal compared to 9 percent of customers who have a low-effort experience, reports Gartner. Disloyal and unhappy customers are also more likely to speak negatively about the brand.
However, everything has two sides to it. By reducing customer effort, companies are also reducing opportunities to interact with the customer and to demonstrate the value they provide. So, in addition to measuring the amount of effort customers are experiencing, it’s important to understand the context behind the interaction. Are you reducing the number of high-effort customer interactions where you are demonstrating your value-add to customers? Are there other ways to communicate with the customer that are less intrusive but helpful in creating a sticky relationship with the customer?
For instance, customers may prefer to delete an account or end a subscription without speaking to a representative, but it would be helpful to understand why the customer is churning. Instead of forcing customers to speak with a representative, have them fill out a short exit survey. Measuring customer effort is only one piece of the puzzle. Make sure the other pieces are included to get a full picture of the customer experience.
2. Getting emotion analytics right
There are a lot of things to like (pun intended) about customer emotion analytics. Such analytics can provide a better understanding of customer behavior. For example, customers who are angry may generate revenue at a slower rate than those who are happy. We’ve also seen instances where customers rated a client’s service with positive satisfaction scores, but upon closer inspection using emotion analytics, the customers were actually frustrated because only the last associate of several that they spoke with was able to resolve their issue. Looking at satisfaction in isolation can fail to capture how customers really feel.
Customer emotions can also help companies make more informed decisions. If a beta release is sparking angry responses from the test group, the company knows that it needs to make changes. Customers’ emotional reactions to an experience have also had more impact on customer loyalty than how effective or easy their customer experience was, according to Forrester.
But customer emotions are only as useful as the system for mining the data and acting on those insights. Emotion detection and analysis can leverage a combination of words and rich voice data among other things. Does your company have the resources for parsing through the data and deriving insights that associates can act on?
Additionally, emotions are fluid and complex, making them difficult to pin down. A customer may seem cross but it’s for reasons unrelated to the company. Emotion analytics can provide a new facet of the customer experience, but companies should proceed carefully.
3. Speech analytics is all about the methodology
Speech analytics, particularly query-based speech analytics, are widely used by companies to mine insights from live calls, chats, etc. Speech analytics are helpful for identifying patterns and key words in customer interactions.
However, most speech analytics deployments won’t tell you about anything that you haven’t asked it to look for. The value in query-based speech analytics depends on the categories or search phrases that are built and require an analyst to identify the impactful words and phrases that people (customers or associates) use in conversations, while also making the search terms specific enough to return the sort of useful insights you are looking for.
Unless you are using a data driven, AI-powered approach, you could be missing important opportunities. In addition, queries where there are no specific key words and where context is paramount can take an analyst weeks to build. Leveraging machine-learning models will enable companies to process interactions faster and at a larger scale than a human could.
A double-edged sword
Customer metrics are some of the most effective weapons in a business’ CX arsenal. But like swords, metrics are only as powerful as the skill of the person wielding them. Savvy business leaders understand that truly valuable data insights begin with asking the right questions.
Is your customer experience strategy stuck in practice mode? Take this quiz to get ‘unstuck.’
