There are a number of different ways to measure customer satisfaction. Performing customer satisfaction (CSAT) surveys is one of the most direct and simplest ways to obtain the customer feedback you need, to understand how your customers feel about your brand and your products or services.
For years CSAT surveys have been one of the industry-standard measures of general or departmental/feature-based customer happiness and loyalty. They usually consist of just a single question that is rated on a number scale, with all responses calculated to give you your CSAT score or percentage.
But with new feedback gathering tools and new types of customer feedback sprouting up daily (social media, online reviews, chatbots, etc.), are customer satisfaction calculation scores still relevant?
Yes. CSAT scores are still a relevant industry-standard metric of customer loyalty and satisfaction with a brand, or its products and services. CSAT scores of customer service and overall customer experience should be regularly compared against industry benchmarks.
There are a number of different CSAT surveys used to measure slightly different areas of customer satisfaction. The three major types of customer surveys are:
On a scale of 1 to 10…
From 0 to 10, how likely are you to recommend [our brand, product, or service] to a friend or colleague?
How easy was it to solve your problem today?
CSAT surveys are still relevant because they’re easy to implement and easy to calculate. They can pop up in your app, in an online chat, on your website, via text, or on in-store kiosks. Because they only ask a single question, response rates are high: customers click the number of their choice – just as simple as closing out of the app or webpage.
CSAT surveys are calculated by percentage – a metric that everyone understands – so, they are easy to relate to, and it’s easy to compare one company’s CSAT score against another's.
If you’re calculating a CSAT score on a scale of 1 to 10, simply add up the overall score given by customers, divide it by the overall possible score, and multiply by 100.
So, if you asked “On a scale of 1 to 10, how was your experience today?” to five customers and received these results:
|Customer||Response Score 1 to 10||Maximum Possible Response|
You’re CSAT is 70%.
If you are working with dichotomous or binary questions that have only Yes/No, True/False, Positive/Negative, etc. responses, you simply divide the positive responses by the total number of responses and multiply by 100.
Let’s say you asked the question: “Did we help you with everything you needed today?” to 200 customers, with 140 saying “Yes” and 60 saying “No”:
Your CSAT score is 70%.
To calculate your NPS score, you subtract the percentage of Detractors from the percentage of Promoters (Passives are excluded). So, if 20% of respondents are Detractors, 30% are Passives, and 50% are Promoters, the NPS score is 50 - 20 = 30.
Because CSAT surveys are flexible and easy to implement, you can ask them at any point of the customer journey: when they first discover your product or service, directly after purchase, after onboarding, after account cancellation, etc. This allows you to dig into different aspects or different departments of your business. They help pinpoint specific issues across your business.
Plus, a customer satisfaction calculation is so quick and the information so easy to gather, you can perform CSAT calculations daily and follow your score as it rises or falls.
A CSAT score of 80% is considered the gold standard for good or very good, although it varies by industry and area of the business. And many industries have seen a steady decline in CSAT over the past three years. (See below for stats by industry.)
The average CSAT for American companies fell 1.2% from Q2 to Q3 of 2020, to 74.4%, the largest quarterly drop in almost two decades, according to the American Customer Satisfaction Index (ACSI). This is, undoubtedly, due in part to the COVID-19 pandemic, connected to lower wages, lower consumer spending, and decreased GDP.
According to a new Zendesk survey, however, 50% of customers worldwide say that the customer experience is more important to them now than it was a year ago. Half also say they would switch to a competitor after just one bad experience, and 80% would jump ship after more than one bad experience.
Measuring CSAT can help pinpoint areas that need improvement, to cut down on customer churn and increase retention. CSAT scores are designed to give a general view of customer happiness. They shouldn’t be viewed as a comprehensive rating of success, but rather, a tool to compare your business against industry standards and competitors working in the same field.
An NPS score above 0 is baseline “good” because it means you have more Promoters than Detractors. An NPS of 20+ is considered satisfactory to favorable; 50+ is excellent, and 70+ is among the world’s premier companies. No company has yet to score an NPS of 100.
A good CSAT response rate is anything around 25%, and 50% or higher should be considered an excellent CSAT survey response rate. The average across all companies and industries is about 15%. Some companies have rates over 40%, but those tend to be businesses with overwhelmingly happy customers, so it’s actually not as useful. Even though it can be painful to read, in general, negative feedback is much more helpful.
How many customers respond to your surveys can depend on how you distribute them, how easy they are to complete, and the demographics you’re surveying. Sometimes incentives can entice more customers to complete them and make sure your survey invitations and directions are clear and straightforward.
As we discussed earlier, CSAT scores and NPS scores are still very much relevant, just not entirely comprehensive. Comparing your company’s CSAT against your competition, or the industry as a whole is where they’re often most useful.
|Industry||CSAT Average||NPS Average|
|Wireless Phone Services||74||30|
|Internet Service Providers||65||-7|
9 strategies you can implement right away to improve your CSAT score and increase customer loyalty:
As mentioned above, you could lose half of your customers to a competitor after a single bad experience. Better customer support requires modern techniques.
Customer support is quite often the largest pain point in the customer journey for any business. With the help of AI tools in customer care, you can monitor customer support data (from customer support tickets, CRM systems, emails, chatbots, and more) and automatically analyze it for major insights. Follow your support team’s success and uncover the major pain points with almost no human interaction needed.
MonkeyLearn is an AI-driven text analysis platform with powerful tools to translate customer service data into action, saving you time and money, and producing much more accurate results than human analysis.
Customers expect their support tickets to be handled fast, especially when service stoppages or bugs keep them from accomplishing their goals or shut down their work completely. You may need to hire more employees or retrain your staff for best results.
Or put automated text analysis to work on your support tickets. Categorize them by subject or topic and automatically route customer support tickets directly to the proper department or employee – even analyze them for urgency, to make sure the most important matters get handled first.
Simple phone or email support isn’t enough these days. You need to meet your customers where they “live” – with in-app chats, on your website, on forums, social media, etc. It’s all about making support convenient for the customer. Surveys show that 18% of customers who comment or complain on social media expect a response immediately, while 83% expect it to come the same day.
According to the Zendesk CX report: “Data shows that companies who perform better across key CX metrics, including faster response times and higher customer satisfaction rates, are more likely to have adopted multichannel support.” The report shows that over 50% of high-performing international companies offer multichannel support
You need to open all of the potential communication channels and monitor each one with AI tools. Twitter sentiment analysis, for example, can constantly monitor for tweets related to your business, then organize them by subject and opinion polarity (Positive, Neutral, Negative), so you can manage complaints all over the web in real time.
Give your staff the power to make their own decisions and resolve customer issues without needing to push them through the chain of command. A properly trained staff should be able to handle their own issues – which saves time, builds trust with your staff, and improves their work pride and job satisfaction.
Open-ended feedback is customer feedback that can’t just be answered Yes/No, on a number scale, or with multiple choice questions. Open-ended survey questions lead to responses in the respondent’s own words, offering ideas, opinions, and feelings, rather than simple quantitative data that is easily represented as numbers, percentages, and statistics.
Machine learning techniques, like sentiment analysis of survey responses, can automatically process thousands of open-ended responses to help you understand which aspects of your company are most positive and which are most negative.
A customer feedback loop is a system to (1) gather customer feedback, (2) analyze customer feedback, and (3) act on customer feedback. It’s a foolproof technique for ensuring that you’re always listening to your customers and putting their feedback to good use. Feedback loops use CSAT scores and other, often open-ended, feedback data to follow and understand the whole customer journey.
Once you have analyzed and applied your customer feedback, it’s important you “close the loop,” and let them know that you have implemented changes, or at least acknowledge that you have received their comments. Sometimes just letting your customers know you’re listening can be a huge step toward building lasting relationships.
One simple way to make your customers know that they’re at the center of everything to do is to show them that you actually know who they are, individually. Integrations with CRM software or email and chat applications can help ensure you always address your customers by name, no matter the form of communication.
Text analysis techniques, like named entity extraction can extract names of people and organizations, dates, addresses, etc., to automatically populate spreadsheets for later use, or format them for use with other apps and programs. Sending messages and promotions on milestones, like birthdays or service subscription anniversaries can help remind your customers that they’re always on the brand’s mind.
There’s nothing more frustrating than walking up to a deli counter, for example, and the employee, no matter if they are presently occupied or not, won’t even acknowledge you. Sometimes a simple, “I’ll be with you in just a minute,” can help calm the anxiety of waiting to be helped.
When you receive customer tickets, it’s important to respond right away and set expectations of wait time for your customer. Even if it’s going to be a long wait, it’s best to let them know right away. And, maybe, more importantly, be sure you’re judging the timing correctly. Don’t tell customers that you’ll be able to help them in a few minutes, if it could be up to an hour – it will only make the problem worse.
SaaS tools, like MonkeyLearn, can help you analyze customer feedback about your brand on social media, online reviews, forums, news articles, and all over the web. Techniques like aspect-based sentiment analysis, process feedback about your brand, organize it by category (e.g., Pricing, Usability, Features, Shipping, etc.), then analyze it for sentiment (Positive, Neutral, Negative), so you understand the categories or “aspects” of your brand or product which are most positive and which are most negative.
Furthermore, unsolicited feedback is often the most useful because spontaneous comments are often the most honest and offer the deepest, most heart-felt opinions.
Once you’ve collected your CSAT and NPS data from close-ended surveys, you can use desktop processing tools like Google Sheets and Excel for data analysis. Find out where your company is performing best (and worst) by demographic or geographic location. Follow your CSAT scores as they rise and fall over time, and understand which particular features of your product or service may need improvement.
When you’re ready to dig into the customer experience even further with open-ended questions and qualitative data analysis, take a look at MonkeyLearn. MonkeyLearn’s suite of text analysis tools are powerful, easy to use, and can be put to work on virtually any kind of text data: surveys, emails, chatbot conversations, news articles, online reviews, social media, and more.
January 29th, 2021