Happy, successful customers are the lifeblood of any business. They’re what transforms your growth from a funnel into a flywheel. You can create happy customers, but to do it you have to truly understand them and how they think about you.
So how do you know what they’re thinking? You have to ask!
My name is Michael Redbord, and I helped build HubSpot's support and service teams, working with customers for over eight years. Today, I’m the General Manager of HubSpot's newest product line, Service Hub, which helps our customers grow through exceptional service.
At HubSpot, we use feedback to understand our customers and what the market demands. This approach has driven dramatic changes over the years—taking us from a small marketing app to the more complete front-office software suite we are today.
Today, I’m sharing the most important things I’ve learned about customer satisfaction surveys, so you can take some of my learnings and use them to adopt a practice of customer feedback at your business and grow better by mastering your market.
A customer satisfaction survey is a questionnaire designed to help businesses understand what their customers think about their products or services, their brand, and their customer support. Customer satisfaction surveys allow companies to improve products strategically, optimize user experience, and deliver exactly what the market demands.
At HubSpot, we use customer satisfaction surveys to understand the micro-level experiences of individual users and address their concerns, but we always return to macro-level questions like “Where is the market going,” “How is our product on the cutting edge,” and “Where do we need to improve?” as well.
4 types of customer satisfaction surveys
There are several ways to understand if your customers are happy, loyal, and satisfied with your product or service, and each can help you learn something about your users' experience. Here are four of the most effective surveys you can use to understand your customers.
Customer Satisfaction Score (CSAT)
Customer Satisfaction surveys usually contain a simple question with a binary response (e.g., yes/no, happy face/sad face). They ask things like “Did our product do what you wanted it to do?”
These scores are usually high (in the 98%+ range), so a sudden spike in negative scores tells you there’s an issue that needs immediate attention.
Net Promoter Score (NPS®)
A Net Promoter Score survey asks customers to rate how likely they are to recommend your company/product to a friend or colleague on a scale of 0-10. You then compare your percentage of detractors (0-6 answers) to that of promoters (9-10 answers) to see where your company stands—the more promoters you have, the more you can infer people are satisfied with you.
Customer Effort Score (CES)
The Customer Effort Score measures how much effort it takes for customers to either use your product or fix a problem through customer support.
At HubSpot, we send a CES survey after we close each support ticket. A CES question will ask “How easy was it to solve your problem?,” and customers are typically given a 5-point scale with options like ‘very difficult’, ‘somewhat difficult’, ‘about as easy as I expected’, ‘somewhat easy’, ‘very easy'.
Milestone surveys are questionnaires sent out at key moments in the customer journey to help understand the user experience better. A milestone can either be time-based (e.g., sending a survey 60-days after signup) or experience-based (e.g., sending after onboarding is complete).
Creating your customer satisfaction survey: 6 types of question
There are different ways to ask customer satisfaction survey questions, and each has advantages and disadvantages. Here are six types of questions you can ask.
In multiple-choice questions, there is a limited number of answers a respondent can give. The results are easy to tabulate, and you're more likely to get a response from a user because these kinds of questions require less effort to answer than open-ended ones (where users need to type in their answers).
Multiple-choice questions include rating scale questions, binary scale questions, nominal questions, Likert scale questions, and semantic differential questions, which are all explained below.
Rating scale questions (also called ordinal questions)
Rating scale questions offer a range of multiple-choice answers that map onto a numeric scale, such as rating customer support on a scale of 1-5 or stating the likelihood of recommending a product on a scale of 0 to 10 (NPS question).
How likely are you to recommend [product name] to a friend or colleague? (0-10)
Binary scale questions
Binary scale questions limit respondents to one of two possible answers, such as ‘yes/no’, or ‘thumbs up/thumbs down’. They work well to limit the ambiguity that comes with subjective responses to rating scale questions (because one person’s 4-star experience is another person’s 5-star experience).
Did customer support help you find the solution you were looking for today? (yes/no)
Nominal questions identify different categories of answers. The answers don’t overlap (unless you include an ‘all of the above’ option), and you can’t apply a numerical value to them. Example
Which of the following bests describes you? “I am ______”
A current customer of [Company Name]
Not a customer, but considering becoming one
Not a customer and don’t plan to become one anytime soon
Likert scale questions
The Likert scale is a 5- or 7- point scale used to evaluate how customers feel about something. The bottom number (always a ‘1’) represents one extreme view, while the highest number (e.g., a ‘5’ on a 5-point scale) represents the opposite extreme view. The middle number (e.g., a ‘3’ on a 5-point scale) indicates a more moderate view.
To what extent do you agree with the following statement:
[Product name]’s onboarding process was simple, straightforward, and painless.
1 - Strongly disagree
2 - Somewhat disagree
3 - Neither agree nor disagree
4 - Somewhat agree
5 - Strongly agree
Semantic differential questions
Semantic differential questions are similar to Likert scale questions in that they both use a 5- or 7-point scale. What makes semantic differential scale questions unique is that they are more descriptive and ask the respondent to choose the option that best represents their opinion or attitude on a given subject instead of asking them to simply agree or disagree.
‘Was there anything in your checkout process we could improve? If so, what?’
11 (specific) questions to ask on a customer satisfaction survey
Question 1: How would you rate the support you received?
Question type: Customer Satisfaction (CSAT)
CSAT surveys usually ask respondents to pick from two options, and scores tend to be high (98%+) so a sudden drop can alert you to problems.
Question 2: How happy are you with [product name]? (1-5)
If they answer 1-3, ask ‘Sorry to hear that! How could we improve?’
If they answer 4-5, ask ‘What do you love about [product name]?’
Question type: milestone survey
Send out milestone surveys at key points in the customer lifecycle, such as mid-way through their onboarding, after several months as a customer, after one year as a customer, etc.
Question 3: How likely are you to recommend [product name] to a friend or colleague? (0-10)
How likely are you to recommend [product name] to a friend or colleague? (0-10)
Question type: NPS survey question
Net Promoter Score (NPS) surveys are best used after customers have had a chance to experience your brand and form an opinion. At HubSpot, we wait 90 days after onboarding to send our first NPS survey.
Question 4: How easy did we make it to solve your problem?
1 - Very difficult
2 - Somewhat difficult
3 - About as easy as I expected
4 - Somewhat easy
5 - Very easy
Question type: Customer Effort Score (CES)
At HubSpot, we ask a CES question every time we close a support ticket. It helps us understand not just the customer’s experience with the ticket itself, but with the whole support experience.
📝 Editor’s note:
Questions 5 to 11: more customer satisfaction survey questions
Here are seven additional questions that work well for customer satisfaction surveys, taken from Hotjar’s list of survey questions for different industries.
How could we improve your experience with [Company Name]?
Do you have anything else to add? Please be 100% honest; we love feedback!
Please describe yourself in 1-2 sentences. For example, "I am a 35-year old data analyst who lives in San Mateo, California.”
What should we do to ‘WOW’ you?
How would you feel if you couldn’t use [Product Name] anymore?
What would persuade you to use [Product Name} more often?
What, if anything, almost stopped you from purchasing from us?
How to create and distribute a customer satisfaction survey
Getting started with customer satisfaction surveys can be hard, but here are some steps to make things a bit easier.
Step 1: start small
If you’re new to collecting feedback, you don’t need to set up all of these surveys at once. You can (and should) start small, testing out a few survey questions and seeing what kind of response they get. Over time, you will have opportunities to refine your technique.
Step 2: decide what to measure, how to measure, and who to survey
What to measure: refer back to the list of the different types of customer satisfaction surveys (NPS, CSAT, CES, etc.) and decide which of these metrics will be your focus.
How to measure: you have a number of options for sending surveys. You can…
Use an on-page survey that pops up in a window at the bottom of the screen
Send an email with the survey attached
Install a feedback widget where users can leave comments on any page
Who to survey: do you want to study new customers? Long-time customers? Renewals? Non-renewals? Those who recently completed a customer service call? The more intentional you can be about choosing your surveys audience(s), the better.
📝 Editor’s note: at HubSpot, the team built their own tools to collect customer feedback. You can either build your own like they did, or you can use online software like Hotjar’s Survey tool (for on-page surveys) or Incoming Feedback(Hotjar’s feedback widget that lets users comment on each page).
Step 3: pick an ideal time to distribute your survey
When it comes to sending surveys to your customers, you will need to figure out what timing makes the most sense for your business.
For customer support surveys, we like sending them 20 minutes after every interaction. This helps us address any issues with the support itself, and the short delay gives customers a chance to figure out if the solution actually worked.
You can also send surveys at different stages of the customer lifecycle, such as:
midway through onboarding
every six months
when you lose a deal (to figure out why)
during renewal, or
when a customer decides to cancel.
Again, make sure your timing is ideal. For example, sending an NPS survey too early could result in bad data since your customers haven’t had a chance to fully evaluate your product yet.
Step 4: evaluate your data once you’ve got a large enough sample size
Use a sample size calculator to determine how many responses you need to have statistically significant data (i.e., a representative sample of your customer base).
Get inspired by these 7 customer satisfaction survey examples
Nothing beats real-world examples, so here are a few from Hotjar and HubSpot.
1. Point-of-conversion survey (Hotjar example)
Who: new customers
When: within seconds of purchase
Where: on-page survey
Hotjar’s point-of-conversion survey appears seconds after a user signs up for the paid plan. Here’s what it looks like:
A point-of-conversion survey helps shed light on any negative experiences that could have prevented the customer from making their purchase. Ideally, you can use these responses to address possible pain points and improve the user experience for future users.
2. Customer effort survey for customer support (HubSpot example)
Who: customers who have requested support
When: 20 minutes after closing a support ticket
Where: HubSpot’s Support Inbox
At HubSpot, we send a Customer Effort Score (CES) survey 20 minutes after closing a support ticket. This delay gives customers a chance to figure out whether their problem was really solved.
You’ll notice that we give them the option to reopen the ticket, which addresses the micro-level experience of the individual user. Next, we ask them to rate the effectiveness of the customer support agent, and then we ask the user how much effort they had to apply to get their problem solved.
Asking these two questions side-by-side helps us distinguish between the customer's evaluation of the support agent (which could point to a training issue) and their experience with the problem itself (which could point to a product issue).
3. Retention survey (Hotjar)
Who: customers who have chosen to downgrade
When: immediately before a customer downgrades
Where: Hotjar dashboard
Hotjar uses a retention survey to figure out why a customer downgraded from a paid plan.
These surveys are a great way to identify and address product or service issues that affect customer retention.
4. Retention survey (HubSpot)
Who: customers who have chosen to cancel
When: immediately after a customer cancels
Where: on-page survey
Here’s how we do our retention survey at HubSpot:
5. Mid-onboarding CSAT survey (HubSpot)
Who: customers in the middle of onboarding
When: half-way into the onboarding process
At HubSpot, we use a simple CSAT survey to make sure we’re handling the onboarding process well.
The red and yellow faces trigger contact from customer support (micro-level focus), and any issues that come up will be explored at a broader level (macro-level focus).
6. Customer effort survey for product use (Hotjar)
Who: new customers
When: two weeks after signup, when the customer has started using key features
Where: on-page survey
You can use a CES survey to evaluate how difficult it is to use your product. Here’s how Hotjar does it:
Notice that both negative and positive responses lead to open-ended questions that help Hotjar improve the user experience.
7. Net Promoter Score (HubSpot)
Who: both new and repeat customers
When: once customers have had a chance to use the product and form an opinion
Where: on-page survey
NPS measures how likely your customers are to recommend you to a friend or colleague. Here’s how we do it at HubSpot:
The results are in: 6 follow-up steps for your customer satisfaction survey
Now that you have all the information from your surveys, what do you do with it?
Step 1: thank respondents
This is a critical step that helps ensure customers will continue to provide feedback. Customers are more likely to give feedback if they know you’re paying attention and value their efforts.
Step 2: close the feedback loop
Follow up with any customers who left negative responses. This is the start of the micro-level focus, where you start working to address individual concerns.
Step 3: read and analyze comments
This is the most important step, and it provides an opportunity to address customer satisfaction on an individual (micro-level) as well as on a larger scale (macro-level).
Step 4: address the micro-level issues
At HubSpot, we have a Slack channel called 'NPS Alerts', where we upload the results from every single NPS survey we receive (score + comments). More than half the company is on that channel, and when an issue is highlighted, we work together to solve it. Once it's fixed, someone will reach out to the customer and CC anyone else who wants to join the conversation.
Step 5: put out any fires (i.e., pull the Andon Cord)
There’s a concept in Japanese car production (Lean Manufacturing) called the ‘Andon Cord‘. If things aren't working correctly in a factory setting, people could get hurt. With that in mind, any employee working in one of those factories can pull this cord, called the Andon Cord, which stops the entire line of production.
Now, at a Software-as-a-Service (SaaS) company, no one is in any physical danger—but when our feedback shows that something is really wrong, we can ‘pull the Andon Cord’ to stop everything and try to figure out what’s going on.
Step 6: address the macro-level issues
When we look at data at HubSpot, we start by ruthlessly segmenting (i.e. dividing users into groups). Say we want to improve the experience for marketing executives on a given product. We start by only looking at their feedback and we go through an exercise to understand, itemize, and rank those comments in order. This allows us to really get into the heads of a particular segment of our users, and we believe it helps us get traction that we couldn't get if we chose not to segment.
Final word of advice: by following these steps, you will be able to build a solid foundation for understanding your customers, addressing their needs and concerns on an individual or micro-level, and improving their overall experience on a larger scale.
Of course, you might not see an immediate increase in your scores when you start making improvements, but if you collect solid data and trust in the process, your scores will improve over time—and your satisfied customers will ultimately drive your success.
Editor's note: we added the following section to Michael's original article to help you get started with your customer satisfaction surveys.
5 customer satisfaction survey templates to jump-start your customer satisfaction survey
To get started right away, we created and shared five customer satisfaction survey templates you can follow.
🔥How it works: click on each image to see a live version of the survey; you can interact with each to see how they work—if you submit your feedback, we'll see it! Each survey will also direct you to a Google Doc template where we listed all the relevant questions, so you can simply copy-paste them into your own survey as they are or tweak them as needed.
1. Net Promoter Score® (NPS) survey template
2. Customer satisfaction (CSAT) survey template
3. Customer effort score (CES) survey template
4. Point-of-conversion survey template
5. Retention survey template
4 other ways to understand your customers
User personas are semi-fictional characters based on real data about the people who use your website. Getting to know your users will help you stay on top of the market by giving them what they want.
Market research is a collection of techniques used to understand your target market. Good research identifies customer needs, fears, drives, and frustrations. You can use this information to design better products and improve the customer experience.
Hotjar Heatmaps are a visual representation of user behavior that shows where users click, tap, and scroll. They help you understand what page elements are being interacted with or ignored, so you can start forming a clearer idea of what works, and doesn’t, on your website.
Hotjar Session recordings can show the anonymous activity of individual users, which you can tie to their survey answers. In other words, if someone tells you they’re having a hard time using your product, you can watch a recording of their activity to gain insights about their experience.
Have you ever used customer satisfaction surveys to better understand your market? Tell us what you learned and how you used the data to improve your business.
Net Promoter, Net Promoter System, Net Promoter Score, NPS and the NPS-related emoticons are registered trademarks of Bain & Company, Inc., Fred Reichheld and Satmetrix Systems, Inc.
Michael Redbord built HubSpot's support and service teams, working with customers for over 8 years. Now, he is the General Manager of HubSpot's newest product line, Service Hub, where he takes learnings from his experience to help HubSpot customers grow better with exceptional service.