Customer satisfaction

How do you measure customer satisfaction?

Marc Thomas

This guide will walk you through some of the ways that people measure customer satisfaction including surveys, testimonials and data-driven methods

Look, doopoll is a survey tool and so of course I’m going to major on surveying in this guide. But I hope that by the time you’re reading this you’ll also have appreciated that you need to take a rounded view of customer satisfaction measurement, analysis and improvement.

That’s because businesses are complex. Sometimes, I’m sure you’ll agree, you feel like you’re playing whackamole. You knock one issue down and then another pops up.

So it’s not simply enough to take survey data and say things are going well. Instead, we should look at a number of data points for measuring customer satisfaction.

Let’s get surveys out of the way on this list because honestly it’s one of the most effective modes for measurement that you can get when they’re done well and the results are properly treated.

But then we’re going to look at some less obvious solutions too!


Surveys can help us get quantitative indicators of how we’re doing on customer satisfaction. We can definitely get qualitative things out of them too but where they’re excellent is with collecting massive amounts of data and helping us make sense out of that in a structured way.

Warning: plug for doopoll coming here…

We built doopoll to make it as easy as possible for people like you to get responses from people you care about so that you can focus on driving growth in your business. How do we do that? We achieve a 3-4x increase in response rate for our customers when they switch to doopoll. That is incredible compared to competitors. You can get started free here.

OK. So. Now that I’ve given you my elevator pitch, let me dive right back into extolling the virtues of surveys.

We can use a number of really established methodologies to track customer satisfaction. They each do different things and they’re all good for different times. And we’ve got templates for them all

Surveys give your customers a place to praise you where you’ve done well and roast you where you’ve not done well. That kind of interaction just isn’t possible through other methods.

Customer satisfaction surveys give you a chance to hear exactly what your customers are thinking and to be really structured with how you want that feedback.


Given that a testimonial is essentially a customer lending their name (and usually face) to a product, there is a pretty good reason to add testimonials into the mix of your customer satisfaction measurement.

Let’s just be specific about that though.

Firstly, testimonials are often given very early on in the customer lifetime. Often, a company will ask for a testimonial during the honeymoon period of product usage. Think about the last time someone sent you a testimonial (or review) request for the product which you had bought. How long after you had purchased did they ask you for it?

There’s a reason that we do that so readily as companies selling a product. It’s because, like a relationship, at first, it’s all rainbows and butterflies. But once the day to day usage of a product sets in, things can look a little less rosey than they did at first.

Secondly, testimonials cannot form the basis of your customer satisfaction tracking. There are a few reasons for this: because there is nothing you can actually measure from them (they’re all qualitative), and also because, as we know, you’re asking them when they’re at their peak satisfaction level (probably).

You should definitely track the number of people willing to give you a testimonial because it’s a good indicator of how people are feeling in the first 30 days or so.

But if it’s all you’re tracking, you’re likely fooling yourself.

Passive data collection

In addition to surveys and testimonials, you can include data driven measurements. Unlike the other two methods for measuring customer satisfaction which we’ve already discussed, passive data collection doesn’t require you to ask a customer how satisfied they are.

Instead, you make inferences about their satisfaction levels based on data that you collect.

For example, let’s say:

  • Like doopoll, you have a software product
  • Like doopoll, you have a 90+% renewal rate on licence each year
  • Then also like doopoll, you can make a safe bet that you have a high level of customer satisfaction.

Of course, this is only one data point. In reality at doopoll, we measure around 10 key metrics across the year. Some examples of things we look at:

  • Actual usage (number of surveys created by a customer, team members added to an account)
  • Number of sessions that a user has each year
  • Lifetime value – how much a customer has paid us over time
  • Number of support requests per customer

These all feed into a customer satisfaction score for us. It’s one which we track internally in addition to the metrics which we track and share externally.

Like other customer driven companies, it’s important that you measure customer satisfaction using a blend of different methods.

Are there any bad methods?

There’s also a few methods that you should probably file away for a ‘last chance’ scenario:

Telephone calls

One method that some companies will use – although this is more the case for service based companies who are a little more involved personally with their customers – is the telephone interview.

How does it work?

After a transaction of some kind is completed, you’ll often get a follow up phone call which will feel more like a chat than a survey. They’ll ask you some questions and make no mistake, at the other end of the phone line, whether it’s by hand or by machine, the answers you’re giving are going into a system called CATI (Computer Aided Telephone Interview).

Effectively, you’re surveying someone with less ability for them to tell you what they really think – most people won’t be as open with their comments when they know there’s someone with feelings at the end of the line.

That’s why I’d never recommend someone conduct customer satisfaction measurement purely through telephone interviews. It’s just not going to get you accurate insights.

Anecdotal feedback

One of the absolute worst ways of driving change in customer satisfaction (or really anything), is by using anecdotal evidence to do so.

What might this look like in practice? Watching someone use your product successfully, asking someone directly whether or not they are happy with their purchase, reporting something someone said to someone you know.

Do those sound like weird things to do? Yes? Well, friend, people are weird and they frequently use anecdotal evidence because it massages their ego easily and comforts them when things are going wrong.

But what those people haven’t appreciated is that customers often need help to give you real insights into how they’re feeling. For example, if you ask me ‘do you like the product’ in person, I’ll probably say ‘Yes, it’s fine.’

Anecdotally, you could tell your boss that I’m a satisfied customer.

But here are some questions that I would have given you insightful feedback on:

  • In what ways is this product difficult to use?
  • Before you used our product, how were you solving your problem?
  • What is the feature of the product that you would prioritise in a list of things to change? Why?

In fact, here’s an extract of Rob Fitzpatrick’s excellent book The Mom Test:

The Mom Test: Talk about their life instead of your idea; Ask about specifics in the past instead of generics or opinions about the future; Talk less and listen more.

If for some reason you decide that you absolutely must use anecdotal evidence for customer satisfaction measurement, then at least follow the Mom Test’s advice in talking to customers. Be specific about the problem your product is solving and how well it is working for your customers.

I’m going to assume that you’ve been persuaded about using established survey methodologies as at least a part of your customer satisfaction work now and so I’ll continue to talk about the specifics of how that works in the next few articles. Right this way.

Easy to use survey builder.
Create as many surveys and questions as you like.
3-4x more responses.
Responses appear in real time.
Try it free