Measuring Customer Experience: Collecting CX Metrics The Right Way
Customer experience (CX) metrics are what a company turns to in order to quantify the success of their interactions with customers, the degree of satisfaction with the service offered, and a customer’s willingness to transact again based on their last couple of interactions. Measuring these throws up data that you can reliably turn to for answering questions about whether or not a product or service is meeting your customer’s expectations and to what extent.
Customer support teams play a critical role in measuring customer experience as they have front row seats to how a customer experiences a company’s offering and also are in a unique position to capture and produce insights into which aspects of a product, process, and service currently delight and disappoint.
This is one reason why, increasingly, the C-suite has been turning to the support function for generating CX insights. To this end, customer support managers are beginning to cast a deeper focus on the three most common CX-related customer surveys – CSAT, CES, and NPS.
Key CX Metrics: A primer
The most common methods of measuring customer experience include in-person interviews and online surveys. While the most amount of value can be milked from the first, an online survey is the easiest, most convenient method of generating top-line CX metrics today.
CSAT, CES, and NPS are three surveys most useful for quantifying a customer’s perception of experience. They also happen to be measurement instruments commonly used by support teams to determine whether they are meeting customer expectations, making their customers’ journeys easier, and building brand loyalty.
1.CSAT (Customer Satisfaction) survey is the barometer by which most customer support functions determine their success. By and large, it measures a customer’s satisfaction with an interaction or support instance, assessed with a question like “How satisfied are you with the service you have received?”. A customer is asked to rate their response on a scale of 1 (least satisfied) to 5 (very satisfied). While it varies from industry to industry, a CSAT score of 85 and above is considered exemplary across the board.
Pros: CSAT is the simplest technique for determining whether a certain service/communication delights or disappoints. It is most relevant at an interaction level and useful for capturing a customer’s emotions ‘in the moment’. The best part is that it can be deployed with every interaction and throughout the customer’s journey. This way, support teams can narrow down their assessment of interaction down to the touchpoint. CSAT is also particularly beneficial for assessing a change in business process and service delivery mechanism.
Cons: CSAT, while capable of capturing key sentiments, cannot capture nuances. Even when stretched to its limits, CSAT tends to reduce interaction to just good and bad, which burdens support managers to resort to multiple more questions for a reliable root cause analysis. Also, with satisfaction being such a nebulous term, it is difficult to arrive at any universal estimates and facts with CSATs alone.
2. CES (Customer Effort Score) is the measure of the cognitive load of a service/interaction, or more simply – how hard was it for your customer to execute their task or intention. It’s captured by way of a survey, wherein a customer is presented with a question that asks them to determine ‘the amount of effort’ it took to, say, achieve issue resolution, on a rating scale of ‘very easy’ to ‘very difficult’. The central premise of this survey is that a company can increase the odds of customer satisfaction as well as customer loyalty by making their products and services easy to use.
According to Gartner Research, 96% of customers with a high-effort service interaction become more disloyal compared to just 9% who have a low-effort experience.
Pros: Like CSAT, CES is one of the easiest kinds of surveys to deploy and use. It works best when deployed immediately after an interaction when the experience is still fresh in the mind of the customer. A positive customer effort score is one of the best indicators of the effectiveness of your SLAs and internal processes. Also, not only is CES a powerful indicator of customer satisfaction but also the strongest indicator of whether or not a customer is likely to transact with your business again. Additionally, CES has cross-functional use cases. It can be used by any customer-facing team, from product development to customer success, to gauge the effectiveness of a process or a feature upgrade.
Cons: Much like CSAT, CES points only to predominant feelings. By itself, it does a poor job of capturing the nuances of a customer’s relationship with a company.
3. NPS (Net Promoter Score) is an indicator of a customer’s willingness to return to a business and evangelize its products and services. It is measured using a single question – “how likely are you to recommend a product/service/company to a friend or colleague?”. Customers are asked to give a rating between 0 (least likely) and 10 (extremely likely) and, depending on their response, the customer is classified as one of three:
- Promoter: Those who respond with a rating of 9 or 10. These customers are most likely to exhibit loyalty as well as promote the company.
- Passive: Those who respond with a score of 7 or 8. These customers are by and large satisfied with the company/product/service but not to the extent of becoming its promoters.
- Detractor: Those who respond with a score of 0 to 6. These customers are least likely to interact and transact with you in the future and may even discourage others from doing so.
The final NPS is determined by calculating the difference between promoters and detractors. Note that the NPS survey can be sent with every transaction and interaction, or used more strategically as a relationship assessment tool; the survey sent once every quarter or year.
Pros: NPS, although newly introduced, is now considered the gold standard for measuring customer satisfaction with a business – at large as well as at a granular level. Because it is used widely now, NPS allows a business to benchmark itself against competition and industry standards. It is also easy to use and extremely intuitive. More importantly, NPS segments customers meaningfully and helps set reliable expectations of customer loyalty and customer base growth. At its best, NPS is a business success metric that helps predict the potential for future revenue.
Cons: Like other metrics, NPS is inadequate for determining reasons for a customer’s predominant sentiments towards a company’s products and services.
Connecting the dots: Taking CX metrics from prescriptive to strategic
A growing number of companies are turning to support teams to collect and aggregate data related to customer satisfaction, effort, and loyalty to understand customer experience better. Sadly, making the right choices about customer experience metrics is not enough. Mostly, because as it is, all popular metrics tend to ring hollow. For CX metrics like NPS, CSAT, and CES to be a trustworthy intelligence for improvement, they must capture and place data in some context of space, time, and causality.
It must be understood that customer experience (CX), just like revenue and growth, is a KPI. And there needs to be a solid organizing principle for collecting and tracking metrics that correspond to this KPI. By themselves, surveys like CSAT, CES, and NPS are disjointed. I’ll go so far as to call them vanity numbers. Sure they offer a strong signal of perceived experience, but for them to help with an experience design-related decision, they need to be a part of a wider metrics funnel. To reveal drivers, correlations, and dependencies, they need some overarching connecting principle.
Here’s one way to collate CX metrics
Once there is a consensus on the CX metrics that matter to the business – satisfaction, loyalty, effort reduction – a customer support leader must cast their attention to determining touchpoints in a journey that contribute most to a metric. For example, in the case of a large metric like satisfaction, it could mean paying attention to smaller operational or productivity metrics like resolution time (FCR and AHT in particular), and relying on regression tools to do this well.
Once a set of productivity metrics are determined, ensure that corresponding surveys and their triggers are appropriately designed. One of the easiest ways to improve the outcomes of a survey like NPS is to layer it with qualitative questions wherever possible. A relevant follow-up question or two, like, “tell us the reason for your score” or “what can we do better” provides a rich repository of insights on motions that drive affinity as well as those that discourage transactions and interactions.
At Freshdesk, we make sure to send a DSAT (Dissatisfaction Survey) anytime a customer rates their experience poorly over a CSAT. This helps us clarify and classify a poor experience into buckets like support quality, process concerns, response time related, etc. And promptly, too.
The big takeaway here is that the best way a customer support leader can increase the odds of producing strategically useful CX metrics is by layering and balancing operational or productivity metrics and experience metrics. Combining CSAT with resolution time, for example, can help predict the degree of correlation between speed of service delivery and customer satisfaction. Churn rate and NPS put together can establish the likelihood of a customer parting ways with a company. Layering CES and retention data can determine motions that drive loyalty. This, with the understanding that all hypotheses you develop about the relationship between operational and experiential data can be tested using controlled experiments.
The long and short of customer experience metrics
Defining a top-line CX metric and then identifying operational/productivity metrics that contribute to its success is the first step in good CX measurement. However, to collect the right productivity metrics, having the right technology partner is key. A robust system that collects and measures operational data across multiple channels and at scale unearths the most amount of value and helps in constructing the most strategic CX plans. Without such technology, you run the risk of leaving key decisions to judgments instead of data.
Illustrations: Raghuram Jaganathan