In this guide
We’ll look at three of the most popular satisfaction metrics, Net Promoter (NPS), CSAT and CES, and explain which to use in which situation.
More importantly, we’ll look at why the metric you pick doesn’t actually matter, and give you three concrete things you can improve today which will lead to a measurable increase in customer satisfaction.
Let’s get one thing straight before we start:
A wise farmer once said,
“You don’t fatten a pig by weighing it.” – 🐖
Measuring customer satisfaction is not the same thing as improving satisfaction.
This is the single biggest missed opportunity in customer service improvement. Worldwide, in every industry, firms realise that satisfaction is important, so try and measure it by gathering customer feedback.
But they gather data about satisfaction in a way which does nothing to improve satisfaction. Or worse, a way that decreases satisfaction.
You wouldn’t measure rooms in your house with a tape measure that put holes in the wall. You wouldn’t check the temperature of a glass of water by pouring mercury into it and hoping for the best, but time and time again, companies try to measure satisfaction in a way which annoys their customers.
Pro tip: That doesn’t end well.
Which is why, if we may be so blunt, if you haven’t got the basics right, it genuinely does not matter which metric you pick. NPS, CSAT, CES, … or something exciting and home-grown.
They’re all decent measures, and we’ll walk you through the pros and cons of each, but if your ultimate goal is happier customers (that is your ultimate goal, right?), none of them will make your customers happier.
So work out how you’re going to do that first.
So why bother picking one at all?
There’s one good reason to measure satisfaction – it makes it easier to talk about.
Once you’re measuring satisfaction accurately and fairly, you can…
- Report to your board whether your initiatives are working, and win their confidence.
- Keep your team focused on a number they trust and believe in
- Try to benchmark against your competitors (although it’s pointless to do so).
So, it’s totally fine to measure satisfaction.
What isn’t fine, is trying to measure it before you’ve got a concrete plan to improve it.
What does make a difference?
We’ve decades of experience in improving customer service.
We’ve seen methodologies, metrics, and fads come and go – ISO 9000, Six Sigma, NPS, Social Customer Service, Agile, Chatbots (remember them?), and now AI – all promising to totally revolutionise customer service.
But just think about your own experience…
…is customer service in general better than it was 10 years ago? Or 20? If any of these things were the ‘silver bullet’ that actually delivered perfect customer service, surely customer service would be great by now?
All of these methods are fine, but they’re just tools to get the job done, not magic that can save a company that’s got the fundamentals wrong.
We want you to be successful – with happier customers and a happier team. So, before we break down which metric to choose, we’ll look at the three fundamental things you need to get right first: Culture, Strategy and Execution. Get them wrong, and you’ll struggle to improve satisfaction – no metric can save you.
Then, we’ll compare the three the most popular satisfaction metrics, and help you work out which is best for you.
A powerful thing happens when you ask a customer for feedback.
Your customer will start to dare to believe that your company cares.
Not whether you care - you wouldn’t be doing it otherwise, but whether there is a goal shared by everyone in your company to deliver great service.
You cannot deliver great service by yourself; it requires everyone in every role to commit. And unless your company culture expects that everyone will deliver great service, customers won’t experience great service.
What’s rewarded in your company? What values are upheld consistently? What behaviours are modelled? What do people get praised for? What’s valued?
Until every leader and everyone has a shared vision and commitment to delivering great service there is no metric, technique or silver bullet that will improve customer service.
You need to fix culture first. Paying lip service isn’t enough, it has to be ingrained in the DNA of the company that better service leads to better financial performance.
This requires belief and attitude, not rules, scripts and tick boxes.
Don’t pick a metric until you can honestly say…
- The leaders in your company are all committed to delivering great service
- Every department in your company (yes, that includes Finance, IT and Sales) are bought into the vision
- Everyone understands that customer satisfaction isn’t a fluffy nice-to-have, it leads to a better bottom line
If you’re fortunate and you’re in a company where that’s all in good shape, then read on.
Great, so everyone in your company 100% understands that happier customers leads to a happier bottom line, and that it’s everyone’s job to make them happy.
What are you going to do about it?
We’re sure the budget airline who send us a 57-page customer survey every single time we travel with them genuinely do want happier customers. And the hotel chain who send us 5 nagging survey reminders every time we use one of their conference rooms are probably keen on the idea of customer satisfaction.
It just feels like… Perhaps they haven’t thought things through.
Culture is great, but unless you have a crystal-clear plan for how you’re going to use customer feedback in your business, it’s better not to bother asking.
Who’s going to feel good about your brand after wading through 57 pages of questions only of interest to your marketing team?
Are nagging ‘reminder emails’ a frontier-pushing customer experience? Or just a way of artificially inflating response rate?
And even if customers do eventually get ground down by your reminders, and slog through your mammoth, boring survey, what are you going to do with the responses?
Because if you’re just hoping to use them in a process improvement project one day, then all the customers who have left you feedback begging you to solve their problems now will have switched to a competitor long before you get round to tackling the underlying issues.
Don’t pick a metric until you can answer these questions…
- When are customers most likely to want give feedback?
- How easy can we make it for them?
- What does each customer reasonably expect to happen after they have given feedback?
Staying with that final point, What does each customer reasonably expect to happen after they have given feedback?
The answer? Customers might want you to respond.
They might not want to be put in a spreadsheet and drawn on a graph. They might have a problem which they need you to fix, and quickly.
If you’re committed to service, you need to be prepared to fix those problems.
Not in “up to 5 business days”. Seriously? If I walked into a bricks-and-mortar store and asked a question, it would be OK to make me stand waiting 5 days for an answer?
Of course not - so it’s definitely not OK to make someone who’s taken time out of their busy day to give you feedback wait that long either.
It’s totally fine to collect feedback and use the results for process improvement, but if you don’t want to decrease customer satisfaction by simply asking for feedback you need to be set up to deal with the responses as they arrive.
(Pro tip: This is one of the reasons why annual surveys are a bad idea)
Don’t pick a metric until…
- You know how you’ll make the whole process of giving feedback a great experience for your customers, especially what happens after they have given it.
- You’ve got processes for short term and long term follow-up on feedback to make sure you increase customer satisfaction instead of just measuring it.
- You’ve put a date in the diary to review and refine how you collect and act on feedback to make sure it is delivering customer and company benefits. It would be odd if you didn’t ask what customers think of your feedback process too!
How to choose the best customer satisfaction metric for your company.
Ok, now you’ve sorted the essentials:
…You’ve guaranteed that your effort to measure satisfaction isn’t going to decrease satisfaction.
It’s time to pick a metric!
We’ll look at the pros and cons of three of the most popular metrics – NPS, CSAT and CES.
Net Promoter Score.
Respondents are then characterised as promoters, passives or detractors according to the score they give you.
NPS is calculated on a scale of -100% to +100%. A positive score indicates that your company will grow, negative means it is likely to decline.
NPS has a few benefits:
- A score out of 10 is simple for your customer
- It’s widely understood and there are lots of products available.
- It’s a strong measure of sentiment. Being prepared to wholeheartedly ‘recommend’ is a tougher test than just asking whether I am satisfied myself.
- NPS is more than a score, it has evolved into a full system, which emphasises the importance of ‘closing the loop,’ that is, following up with customers.
In our experience, if you do NPS correctly, it has few drawbacks. The only significant one is that the ‘recommend’ question isn’t appropriate for all circumstances (but even then, NPS allows you to ask another question if you really, really want to!)
Trouble is, a lot of people don’t do NPS well, and so introduce a few problems. Here’s some common ones we see:
- Using NPS as a customer-facing score. Most customers don’t know what it is
- Only asking the NPS question. Customers want to answer customer-centric questions. Make sure your NPS survey also asks questions they care about, rather than just the ‘recommend’ question which you care about.
- Focusing on the number – it’s great that you have your Net Promoter Score to simplify conversations around satisfaction, but it’s important to not obsess about that score. Instead, obsess about doing things that make customers happier.
Use NPS if…
- It makes sense to ask questions about ‘loyalty’. NPS might not be right if you’re a monopoly, or asking the question early in a customer journey.
- You’re going to commit to the full NPS system, including closing the loop with individual customers.
There’s no one fixed, ‘official’ definition of what CSAT is.
Everyone agrees it’s basically “ask your customers if they’re satisfied”, but after that, all bets are off.
You can use any scale, ask the question in any way, and do anything you like with the results, and still call it CSAT.
In our view, that’s not a massive problem.
As long as you’re doing all the other bits of your feedback process right (asking customer-centric questions at the right time, and following up on the responses), it’s fine to ask customers how satisfied they are.
Here’s a few pointers to help you get it right:
- Don’t get customer feedback mixed up with market research. It truly doesn’t matter what scale you use, so long as customers understand it and can answer it easily. It doesn’t matter if you ‘bias’ the scores in any way, because you’re not trying to do statistics, you’re trying to find problems and fix them.
- Because ‘anything goes’ with CSAT, benchmarking CSAT against your competitors is an even worse idea than ever. They’re probably not calculating it the same as you, and the comparison is totally meaningless.
- Like NPS, don’t just ask the ‘satisfaction’ question. Ask some diagnostic questions, both scored and free text to understand why customers are (dis)satisfied, and to give you an opportunity to fix problems.
Problems with CSAT?
Like NPS, a lot of the criticisms of CSAT (see this article for a few) can be answered with a shrug and a ‘so what’?
- It’s true that cultural bias probably is a factor, the scores are subjective and there might be a bias in that indifferent customers can’t be bothered to fill your survey in, but they miss the point: the accuracy of your score isn’t anywhere near as important as what you do with the feedback you receive.
- That said, Some research suggests that the relationship between satisfaction and loyalty isn’t straightforward. Just because a customer is ‘satisfied’, doesn’t mean they will stick around.
Use CSAT if…
- You want to keep things simple
- NPS isn’t a good question to ask at a given point in the customer journey
Customer Effort Score (CES)
Like NPS, Customer Effort Score was first introduced to the world in the weighty Harvard Business Review.
Its creators, Matthew Dixon, Karen Freeman and Nicholas Toman argue, based on studying over 75,000 customer interactions, that great customer service is very bad at producing loyal customers.
Customers are loyal to great products and powerful brands. Customer service interactions can only do a little to improve loyalty, but can do a lot to destroy it (this probably feels familiar if you’ve ever been on the receiving end of bad customer service!)
So, they argue, if you’re trying to ‘measure’ customer service, it makes no sense to try and measure the ‘best case scenario’ – it doesn’t matter if some customers are delighted if the majority are frustrated. Instead, the metric they propose is more of a ‘worst case scenario’:
“How much effort did you personally have to put forth to handle your request”
(Measured on a scale from 1 (very low effort) to 5 (very high effort)).
CES’ creators claim that it has greater predictive power of future customer behaviour than NPS (and far greater than CSAT).
We like CES because the question itself is customer-centric (no customer wants to put a lot of effort into sorting out problems caused by an organisation). But remember – as with all metrics, make sure it’s not the only question you ask. (Read our guide to customer survey questions if you need help working out what to ask).
Use CES if…
- You’re measuring a single customer interaction, not a relationship
- You’re wary of over-investing in customer service, by spending money on lavish things customers don’t want, rather than just getting the basics right.
Learn to be very clear about why you’re doing customer feedback.
Metrics and process improvement are important, but you’ll struggle to improve ‘customer satisfaction’ unless you improve the satisfaction of individual customers. One by one. Just exactly the same way that you won them as customers in the first place.
So the main goal should not be about gathering data. The main goal should be to check that the experience of every individual customer is excellent every time they deal with you, and to fix it straight away if something has gone wrong.
Improving your processes can then be done just as well, and probably better, as a by-product of the data you collect when you’re achieving that more important goal.
The most important thing to remember is that success is governed by principles that are far more important than which customer satisfaction metric you choose.
These factors have much more impact:
- when you ask for feedback;
- how fast and easy you make it for the customer to give feedback;
- what other questions you ask;
- whether, how and how quickly you respond;
- who in your company gets to see customer comments.
Once you’ve got the basics covered, and only then, how should you decide which metric to use?
- You could just pick a metric you like. Providing you get the more important factors right then almost any metric can be made to work well. If you want help choosing the right one for your needs, just ask us!
- A combination could be best - CES for transactions and NPS for relationships and loyalty.
- CSAT is good if you don’t have a strong preference for one of the others and you just want something that’s easy for your customers and your staff to understand.
Whichever one you choose, you’re now equipped for happier customers, a happier boss, and a happier team!