Once this aim was more of a vague notion than a specific priority, but today customer experience has become the number one focus for business leaders. Here’s what they said when asked for their business priorities over the coming five years:

1. Customer experience (chosen by 45.9 percent of respondents)
2. Product (33.6 percent)
3. Pricing (20.5 percent)

I know what you’re thinking. “Customer experience is more important than my product? Surely that can’t be right?”

This shouldn’t be so surprising. Happy customers are happy for a reason. Maybe you delighted them with a surprise discount, a silky-smooth buying process, or lightning-fast delivery.

In other words, they feel happy because of something you did, and they’re prepared to reward you for it. They’ll pay up to 16 percent more for your product, and they’ll be more loyal to your brand, too.

To improve your customer experience and make them even more satisfied, you first need to understand how happy they are right now.

E-commerce surveys are one of the best ways to do this. With that in mind, here are 9 best practices to help you create effective, insightful surveys.

Want More Conversion Optimization Tips?

We’ve put together 12 CRO resources to help you drive more leads and revenue.

Whether you’re looking for satisfaction guarantee examples or product page examples, we’ve got something for you.

You’ll also get immediate access to 23+ other bonus resources, categorized in Notion for your convenience.

Download Swipe File Now →

Table of Contents

1. Choose the Survey Type That Best Fits Your Needs

2. Get Your Timing Right

3. Start Big-Picture, Then Get Granular

4. Make a Habit of Sending Regular Surveys

5. Cut Out Unnecessary Noise

6. Keep It Short

7. Only Offer a Small Range of Ratings

8. Send a Test Survey Before the Real One

9. Steer Clear of Loaded Questions

1. Choose the Survey Type That Best Fits Your Needs

E-commerce surveys are like a bag of Matador Mix—they come in many flavors. Those flavors include:

• Net Promoter Score (NPS): Customers are asked a single, simple question—“How likely are you to recommend our business to a friend or colleague?”—and asked to give a score from 1 to 10.

• Customer Satisfaction Score (CSAT): Designed to assess a customer’s feeling toward a specific event. Customers are asked to answer, in the form of a rating from 1 to 5, a question like: “How would you rate the quality of your most recent service interaction?” 

• Customer Effort Score (CES): Used to measure how easy (or hard) it was for the customer to complete a specific action. The classic CES question reads: “On a scale of ‘very easy’ to ‘very difficult’, how easy was it to interact with Company X?”

As well as the quantitative data-gathering described above, each survey type can also include a qualitative element in which respondents are asked to explain their initial rating.

Each survey type has its strengths and weaknesses, and the right choice for you will depend on the purpose of your survey:

• NPS: These quick-and-easy surveys tend to generate high response rates and, because they’re quite general, can be used at any point in the customer lifecycle. The results are most meaningful when NPS surveys are sent regularly over an extended period.

• CSAT: If you’re looking to measure the performance of your customer service function, a CSAT survey is the one for you. 

• CES: Looking to identify speed bumps in your path to purchase and smooth out your user journey? A CES survey is your best bet.

Alternatively, you could build a custom e-commerce survey using a tool like SurveyMonkey, where you can ask whatever you want—but you’ll need to do all the legwork yourself.

2. Get Your Timing Right

Timing is everything with a survey. To get the most insightful results, you need to identify the actions that make a customer a good candidate for your survey.

If I’ve not interacted with your brand for six months, it’d be a weird time to ask my thoughts on your user experience. But it could be a good time to dig into why I’ve not purchased from you recently (particularly if I’ve previously been a regular customer).

Clearly, those key actions will vary based on the purpose of your survey, but common touchpoints include:

• After a customer returns an item they’ve purchased;
• After a customer takes delivery of a new product;
• After a returning shopper abandons their basket;
• After a support agent resolves a customer query; and
• After a customer downloads your app.

3. Start Big-Picture, Then Get Granular

Imagine you’ve visited a friend’s house for dinner. They want to know if you’re enjoying yourself, so they ask a bunch of questions: “How’s the lighting? Do you like the wine? Is your seat comfortable? Have you eaten enough? How’s the temperature? Is the music too quiet?”

Then, at the end of the night, they ask: “Did you have a nice time?”

By that point, they’ve already forced you to scrutinize every tiny aspect of the evening, including a bunch of things you might never have considered if they hadn’t asked. That’s definitely going to have an impact—probably a negative one—on your overall verdict.

That’s why you should always start out by asking customers your broadest and most important question, such as how likely they are to recommend your brand, or whether they found you easy to buy from.

Once you’ve asked that big-picture question, dig down into the specifics to provide context for their initial response.

4. Make a Habit of Sending Regular Surveys

Don’t treat e-commerce surveys as a one-and-done activity.

Sure, a single survey will provide you with a snapshot of how happy your customers are at a specific point. But e-commerce is a fast-moving industry, and that information won’t mean much three months down the line.

Instead, send surveys regularly. NPS is only an effective metric if you survey every customer—or at least, a decent chunk of your customers—after they make a purchase. Done regularly, it allows you to track satisfaction over time. Done once, it doesn’t tell you much at all.

There are other benefits to regular e-commerce surveys, too. For one thing, they demonstrate to customers that you’re genuinely invested in improving their experience.

And for another, they let you chart how your actions have influenced customer happiness. If your NPS increases after you switch to a new delivery partner or build a new website, that’s a good sign your efforts are paying off.

5. Cut Out Unnecessary Noise

You probably won’t be surprised to learn that the more questions you include in a survey, the fewer responses you’ll get. 

The lesson here? Don’t get greedy. 

This might sound like an obvious point, but it’s too easy to include questions that you’d like, but don’t need, to know the answer to. Those questions are unnecessarily elongating your survey, which means you’ll end up with less data overall.

Do you really need to know how a respondent first heard about your brand? Will it tell you anything you can’t find out in five seconds on Google Analytics? If you don’t have a watertight reason for including a question, shelve it.

6. Keep It Short

Of course, survey abandonment isn’t solely about how many questions you ask—it’s also about the complexity of those questions.

The same SurveyMonkey study I referenced in the previous section gives us some valuable insight on this. Specifically, it tells us there’s a 10 percent point drop in completion rates between surveys with one open-ended question and surveys with 10 such questions.

That’s because open-ended, free-form questions are harder to answer. It’s not a case of choosing a number from 1 to 5; you have to think—and type—your response.

So how long should your study take to complete?

According to the International Journal of Market Research, the ideal survey takes 10 minutes to finish, and the maximum length should be 20 minutes. Any longer than that and you’ll see a big spike in abandon rates.

7. Only Offer a Small Range of Ratings

You might think a wide rating system is more desirable because it lets your customers give more nuanced responses. But in reality, it could be making your survey harder to complete and ruining your data.

Think about it. Say I’ve asked you to rate this blog post on a scale of 1-3. Easy, right? If you hate it, you’ll score it one. If you love it, I get a three. If you like some parts but hate others, call it a two.

But what if I asked you to rate it out of 10? If you have extreme feelings toward my writing, that’s fine. But how do you choose between a seven or eight? What’s a six?

And more to the point, if I asked 10 other people to do the same thing, how can I be sure they all assigned the same meaning to their scores?

You get the picture. Sure, a broad scale might sound more meaningful, but in reality, it can be exactly the opposite.

8. Send a Test Survey Before the Real One

As with any data-gathering exercise, it’s best to sense-check your survey before rolling it out at scale.

To do that, send it to a select group of customers first and see what comes back. Also, add a few questions about the experience of filling out the survey, or send them as a separate mini-survey.

This approach will help you assess:

• Completion rates: Are you getting enough responses? If not, you may be asking too many open-ended questions, or simply too many questions altogether.

• Data quality: Do the results give you the insights you need? Or is your data too broad or too specific?

• Data type: Are you acquiring enough qualitative data to add meaning to all the numbers?

Don’t be afraid to do this multiple times before you send the first “real” survey to all your customers. These surveys will play a key role in informing future customer experience decisions, so you need to be completely satisfied with the quality of the data you’re seeing.

9. Steer Clear of Loaded Questions

As humans, we naturally want to be loved. We also want things to appear better than they really are. Those two factors can be a major barrier to creating effective and meaningful e-commerce surveys.

You want people to like your brand, so you ask them a question that inadvertently (or deliberately) skews their opinion in your favor.

Or you were personally involved in rolling out a new feature or product, so you highlight how great it is, again influencing the customer’s response.

Leading questions might help you gain higher survey scores. But those scores won’t be valuable, because they won’t accurately reflect how your customers feel about you. That could lead you to make poor decisions based on bad numbers.

What do these deadly leading questions look like? Here’s a great example from SurveyMonkey:

• Leading question: “Our customer support team is rated as being the most responsive in the industry. How responsive—or unresponsive—do you think they are?”

• Neutral question: “What do you think of our customer support team?”

It’s tempting to ask the first question. But if your support team is really that good, you need to trust that your customers will recognize this and score them accordingly.

Want More Conversion Optimization Tips?

We’ve put together 12 CRO resources to help you drive more leads and revenue.

Whether you’re looking for satisfaction guarantee examples or product page examples, we’ve got something for you.

You’ll also get immediate access to 23+ other bonus resources, categorized in Notion for your convenience.

Download Swipe File Now →

Conclusion

If you only take one thing with you from reading this article, let it be this: there’s no such thing as a “bad” e-commerce survey score.

Obviously, you want your customers to be ecstatic all the time, so high scores are the ultimate goal. But a low score is arguably a lot more valuable. If your customers aren’t happy, you must know about it so you can course-correct and shake things up.

So build your surveys, embrace the results (good or bad), and use them to bring about positive change.

How are you using e-commerce surveys today? What’s your go-to survey best practice? Leave a comment.