How product teams can conduct online surveys to get actionable insights

After all, developing a product and improving the user experience (UX) isn’t a cut-and-paste formula: it’s built on iterating and testing small changes, understanding psychographics and personas, and drawing and acting on insights from user data.

This article shows you how product teams can conduct online surveys to get actionable user data by asking the right questions, setting clear goals, and having a framework for analyzing results. We cover:

• What are online surveys?

• 5 steps to conduct a successful online survey:

• Set your survey goals

• Come up with your survey questions

• Choose an online survey tool

• Launch your survey

• Analyze survey responses

Let us begin:

What is an online survey?

An online survey is a virtual questionnaire that provides feedback about a user's experience with your product or service. Generally, brands use data from online surveys to:

• Improve their product

• Better understand their target audience

• Drive their product roadmap

In short, online surveys help you understand your users. Depending on your goals and how you design your survey, you’ll be informed on:

• What users like and don’t like

• What they want, need, and expect from your product or service

• What their goals and challenges are

• How you can change or improve your product to give them a better experience

5 steps to conduct a successful online survey

Every survey can have a unique set of goals (as you'll see below), but all successful surveys share one aspect: they provide you with actionable insights—that is, you can use data from the survey to identify and take steps to directly help users achieve their goals.

Instead of trying to get the highest number of respondents possible, design your survey to provide value to your users: aim to find the highest quality information that will help you help them.

Designed this way, everyone wins.

A successful and high-quality online survey is a thoughtful mix of closed-ended questions, open-ended questions, a decent sample size, but also:

• Design

• Timeliness

• Distribution

• The right balance between getting enough data without overwhelming your respondents

Here are five steps to take to conduct a survey that will give you insights and user data your product team can use:

1. Decide your product team's online survey goals 🎯

Online surveys are often used by customer service teams, to improve the customer experience; and marketing teams, to better serve their target audience. Product teams can use online surveys to improve a product—and better serve your customers.

Since you're building a survey to gather actionable user data, start backwards from your goal. Knowing what you want to accomplish will inform the questions you ask in your survey.

You may have different goals depending on your focus or the stage of your product. For example:

• Product launch

• Product development pipeline

• User testing (of new services, features, upgrades, integrations, etc.)

Having specific goals in mind gives you the power to iterate on specific variables from your survey data and keep your surveys short and to-the-point, increasing the number of completed surveys.

To understand what your survey goals could be, ask yourself:

• What are the questions I want to answer or the problems I need to solve for my users?

• What do I want to learn about the user experience?

• How am I going to use the data I collect from this survey?

  *

* This one is especially important because it will help you understand whether the questions you ask in your survey will lead to actionable insights or not. Think of it this way: the more specific your goals, and the more direct your survey questions, the more actionable your data will be.

Once you've set your goals, it will become much clearer which questions to ask—and who to ask…

Choose who you want to survey (and when)

You may want to target a specific segment of users depending on your goals.

For example:

• Recruit testers for a product change: you may want to use online surveys to recruit more user testers. (Suggestion: offer a prize to get their attention!)

• Users with specific attributes: if you're using Hotjar, use user attributes to target surveys to specific customer segments (e.g. new vs. returning users, paying customers, users with a specific lifetime value, users from a specific industry, etc.).

• Users from certain geographic regions: if you’re localizing your product, you may want to target only users from a specific country or region.

Users based on source: you could trigger your online survey for people who come in from direct traffic rather than from social media.

The timing of your survey makes a difference, too. Alternatively, instead of targeting a specific segment of users, you might want to target users based on when they are in the user journey.

For example:

• Users with specific behaviors: target users who stay on a page for longer than two minutes—or returning users with very low bounce rates.

• Logic jumps: if you’re using Hotjar, you can choose which questions your respondents see based on how they’ve answered a previous question.

• Users who have used your product for a certain amount of time: send an external link survey after a customer has used your product for a week, month, or any specific amount of time.

• Recently churned, downgraded, or canceled users: find out what went wrong and how you can improve the product or UX. (You can even invite customers back after you resolve the issue.)

• Surveys on high-exit pages: sometimes this means people aren’t getting what they want from a page. Placing a survey on high-exit pages makes it easier to identify what went wrong.

2. Come up with your survey questions 💭

There are several types of survey questions you can include in user research. Try a combination of:

• Open-ended questions to get feedback on your site or product in the users' own words, ask open-ended, non-leading questions like “What is the most important feature of our product/service for you?”

• Closed-ended questions: to get quantitative data you can easily sort in a graph or chart, ask questions that allow users to select from a pre-selected, limited number of options—like an NPS survey, which might ask, “How likely are you to recommend our product/service on a scale from 0 to 10?”

• Nominal questions: closed-ended questions for which the answers are non-numerical and don't overlap; nominal questions present the respondent with a set of multiple choices to select from.

• Likert scale questions: these questions check the users’ level of agreement with a statement or the intensity of their reaction towards something, usually on a sliding scale of 1-7 or 1-5.

• Rating scale (or ordinal) questions: rating scale and ordinal questions are similar to Likert scale questions in that they ask users to rate something on a numerical scale. Alternatively, these can be questions where users rank choices in order of importance (like asking them to rank product features in order of preference).

• 'Yes' or 'no' questions: self-explanatory—these questions have your users answer Yes or No.

3. Choose your survey tool 🔥

Now you know why you're surveying your users, who you'll survey (and when), and you know what you'll ask, but…

how do you actually survey your users?

Depending on your goals and audience, some survey tools might work better for you than others.

For example, if you want to keep things plain and simple, a free tool like Google Forms might work for you. But if you want a smart survey tool that is triggered and targeted based on website and user behavior, you might be more interested in Hotjar.

Let's see:

3 best tools for conducting online surveys

1. Hotjar (oh hi! 👋)

What it is: Hotjar's on-site surveys let users answer questions and provide feedback in their own words while they’re on your website, and can be triggered by specific actions.

Instead of sending your survey in an email that users may miss (or ignore), you’ll capture their attention at the right place and time (i.e. when they’re using the thing you want to ask them about).

For example:

• Immediately after the page loads: the survey appears once your site is fully loaded and the Hotjar Tracking Code has been triggered.

• After a delay of x seconds: the survey is triggered after the user has been on the same page for a specific amount of time (of your choosing).

• Before a user abandons the page (on a desktop device): the survey is triggered when the user's mouse moves toward the top part of the browser in any direction.

• When a user scrolls halfway down the page: the survey is triggered once the user scrolls halfway down your page.

There’s always a place for external surveys, though, and Hotjar lets you build external link surveys so you can generate longer, more in-depth surveys too, to dig deeper into the user experience.

Also!, you can customize all Hotjar surveys to fit your branding. 🤝

What it's good for: when you want to connect the dots between user behavior and the user experience.

2. Typeform

What it is: Typeform is a popular data collection tool with a focus on brand and design. It has an elegant interface and lets you easily create attractive branded surveys.

You can use widgets to embed Typeform surveys onto your website or send users a link to the survey. The paid version allows you to use logic jumps for more complex surveys.

What it's good for: design-conscious brands that want to build long surveys and use forms to generate leads.

3. Google Forms

What it is: Google Forms is a great option if you’re integrating data with Google Docs or Sheets. It’s not very stylized or sophisticated, but it works and is simple. Even without the branding and aesthetics that Typeform or Hotjar provide, Google Forms does let you add color themes and a custom logo.

It’s completely free, and you can embed the form onto your website or send out an external link. While simple, it does have useful features, like allowing for file uploads, various question types, and even randomizing question order. It lacks logical jumps and branching but does the trick when you need something (plain and) simple.

What it's good for: budget-conscious survey creators who need a quick and simple survey.

4. Distribute your survey 🚀

Now you need to distribute your survey in a way that gets people to participate and provide you with the actionable data you're after.

How you distribute your survey will depend on your goals and the kind of survey tool you use.

If you use an on-site survey tool like Hotjar, you can distribute the survey while users are interacting with and experiencing your site, giving you valuable, in-the-moment feedback as the experience is fresh in their minds.

For external link surveys, you’ll have to be a little more creative to get your survey out—which is one of the hardest parts about conducting a survey.

Here are a few ways you can distribute an external link survey to your target audience:

• Set behavior-based triggers on your website, for example when their cursor moves anywhere up on the page, signaling they might be about to exit.

• Send the survey to your email list.

• Share a link to the survey on social media.

• Create a unique link for your brand ambassadors and have them share the survey with their network (with incentive—more on this in a second).

• Create a pop-up on your website.

• Create a CTA button that sends visitors to your survey.

• Find user lookalikes on freelancer platforms—like Fiverr or Upwork—and invite them to take the survey.

• Gamify the survey to incentivize your audience, for example by offering them points they can convert to cash in exchange for sharing your survey with their friends.

How to convince people to take your survey

People are busy and don’t necessarily want to take the time to give every brand their opinion about a product or a review of their experience.

Sometimes your audience needs to be convinced that your survey is going to be worth their time. (This isn’t always the case—if your survey has the right balance of 'quick and easy', people may just take it anyway. Yay!)

If you want to incentivize your audience though, it’s important not to offer an incentive so big that you sacrifice quality data.

For example, if each participant gets $200 just for answering a few questions, you might get any random person taking your survey, costing you time to sort through useless data and money spent getting that useless data. (Not ideal.)

Remember that a high response rate doesn’t always mean high-quality data, so be careful when incentivizing.

Here are some ways to convince users to take your survey—and still get actionable feedback:

• Offer a monetary reward (for example, participants are entered for a chance to win something when they take the survey).

• Offer a non-monetary reward like merchandise or a free month with your product.

• Gamifying the survey: get your users to share the survey in exchange for points they can use toward your product.

5. Analyze survey responses 🔍

You made it! Now the last step is analyzing your survey responses.

Depending on your survey tool, you should be able to extract and sort responses in a shareable spreadsheet.

First, analyze quantitative data, because that will help you better understand the qualitative (open-ended) data. Go back to your initial question: what are you trying to understand, or what problem are you trying to solve?

Analyzing open-ended questions does require some focused time, but there’s a formula for doing it (of course there is):

1. Identify response categories.

   These are replies that fall into the same kinds of answers, even if worded differently.

2. Assign categories to individual responses.

   Go through each response and assign the category number.

3. Organize your categories.

   Now that all your categories have numbers and every response is assigned to one, you can group them and identify patterns.

4. Represent your data visually.

   Together with your quantitative data, you can now put your quantifiable data into charts and graphs.

🔥 Pro tip: push Hotjar Survey responses directly to Slack to share with your team so you can process and analyze responses together.

Whenever and wherever you create a survey, toggle on Slack under 'Forward Response' and choose the channel(s) you want to send your survey responses to.

It's a thing of beauty:

Taking the next step in your survey journey

Online surveys are invaluable to improving your product and UX.

Follow the above steps and you’ll be well on your way to getting actionable user data that'll help you identify and take steps to directly help your users achieve their goals.