What makes some website enjoyable to visit and others not so much? Sometimes it’s hard to say but you just know that for some reason website A is better than website B, even if they look similar. It can be clearer navigation, fewer distractors, more appealing color schemes and language that is used etc. All those factors among others contribute to the quality of website experience.
And the quality of website experience is one of the most important factors contributing to the success of a website. If people enjoy visiting it, they will return and perhaps even recommend it to their friends. If they don’t, they will forget about it nad never return. This is why website owners must not underestimate the importance of great website experience.
In order to do it, we read tons of guides (like this one, for example: http://blog.hubspot.com/marketing/improve-your-websites-user-experience#sm.0001txja2p549cpi10zvqb6yaqbu0) and work hard on website design. We want the website to look good, work well on all devices, and be organized in a logical way. But we tend to make one mistake – don’t confront our ideas with the outside world.
We just know too much about our websites – how they are organized, what and how we expect visitors so do. For us, it’s obvious. After days spent on working on websites, we can use them blindfolded. We can easily overlook problems that visitors will quickly notice and which will affect their experience in a negative way. Luckily, it’s easy to find out how visitors assess your website’s experience – simply run a website experience survey and find it out.
What is a website experience survey
Website experience survey is a way of collecting feedback from website visitors to learn how they assess the experience of visiting your website. Website experience surveys are often conducted using website surveys. A small survey widget appears on the website asking for feedback (example below).
Website experience survey template
The first step to running a website survey experience survey is choosing the right questions. It might seem obvious but it is tricky – sometimes changing the wording in questions can dramatically change response rates and answers. Questions presented here work well for Survicate users and for us.
- On a scale 0-10 how do you rate your experience?
Answers to this questions will give you an overview how people assess their experience with your website. It’s the first step to more in-depth website experience survey – answers to this questions are not really actionable. To make it more actionable, ask further questions.
Tip: target people who browsed a substantial number of pages (depending on your website, for some it can be 5 while for others 10) and are about to exit the website. They had an opportunity to familiarize with your website and are likely to know what’s good or bad. If you want to find out why people are bouncing right after entering the website or shortly after, use a simple exit survey.
- What problems did you encounter while using the website?
Find out why people give low grades. You might find out that some links or buttons don’t work, the website loads slowly, or any other reason. However, don’t forget about people who give you high scores.
Tip: ask this question as a second step of the survey after question about the grade. Show it to people who answered between 1 and 7. Use skip logic to do it.
- What did you like the best?
Answers will reveal your most important strengths. You should capitalize on them and not change them if you’re planning a redesign.
Tip: ask this question participants who answered between 8 and 10 in the first question.
- What should we do to improve your experience?
Don’t guess what visitors want, just ask them. This question will provide you with the most interesting insights. One caveat – since this question requires some effort from visitors, response rates are usually lower compared to those to less demanding questions, like the previous ones. You’re likely to see that over 20% of participants don’t finish this step.
Tip: don’t forget to set a ‘Thank you’ message after the last question.
This simple template will provide you with an overview of how people assess your website, what frustrates them most, what they like best and what are their ideas for improvements. But many more elements contribute to the overall experience and if you really want to know what problems visitors face and how you can improve the website you should dig deeper.
One of the most important factors contributing to website experience is navigation (read ‘Don’t make me think’ if you don’t believe me). That’s why this element is worth taking a closer look, especially when you run an ecommerce or news website – they tend to have hundreds of pages and organizing them well is a big challenge. Take a look at this article to learn how to run a good website navigation survey: https://survicate.construction/website-survey/navigation-survey/.
When you start the first satisfaction survey, relax and wait for data.
What to look for while analyzing data
A word cloud will be useful for you for analyzing answers to the last open-ended question – it will show you which words are the most common with answers. But it’s just an overview and you should look at single answers, especially to questions asking about problems users face. Quite often analyzing 30-50 answers is enough to spot the most serious problems. Not so hard, right?
There’s just one small catch – website experience surveys are a bit different from other surveys since the experience can vary a great deal depending on the technology visitors are using. Browsers and devices are the most important variables.
I believe I don’t have to tell you that usually Interner Explorer causes most of the problems but there’s a lot of browser-specific issues. An example from our experience: after updating the design of our website, some of the buttons didn’t work on certain versions of Firefox browser only. We spotted it quickly but if unnoticed, this flaw would have cost us signups and customers – Firefox is the second most popular browser used by visitors to our website.
How to spot similar issues while analyzing results of your surveys? Filter answers based on browsers. You might see that most of the answers suggesting a certain problem come from visitors using the same browser and people using a different browser rate their experience well. Make sure to fix elements that survey participants complain about. Also, test your whole website on this browser as there are likely to be other bugs as well.
Mobile website experience
Researching mobile website experience is not much different from a desktop version. You can use the same set of questions and tips presented in the next section apply to mobile experience surveys as well. The one big difference concerns targeting – you can’t use exit intent targeting because it’s simply impossible. There is no way to detect when a person leaves your website so no survey is triggered. For this reason, it’s recommended to run a specific survey targeted on mobile devices. Set the survey to appear when a person browses more pages than average.
Similarly to analyzing results of desktop surveys, filter answers based on devices or browsers to spot flaws that exist only in certain circumstances.
What’s the fastest way to ruin website experience? Run a website experience survey badly. This includes using intercept surveys – they disturb visitors and for some people seeing such survey is enough reason to leave the website. How to create a survey that doesn’t decrease the quality of website experience and provides reliable results?
- make sure the design of survey widgets is appealing. That said, widgets must be visible – no one will answer a survey if they can’t see it because it matches the design of a website so closely.
- use targeting to adjust the message to the context. Example: don’t ask a question ‘How would you rate your overall experience?’ right after a person enters your website. Trigger this question when a person is about to leave the website after browsing at least a few pages.
- run separate surveys for desktop and mobile devices to adjust targeting.
- use sampling if you have thousands of visitors a day. It might sound strange but it’s better to have 100 text answers to analyze than 100 thousand unless there are people responsible solely for analyzing data. If there’s just one person responsible for the website and he or she is supposed to analyze 20k answers a week, they won’t do it anyway or will look at just a small chunk of data. It’s better to collect and analyze 50 answers than to collect 10k answers and analyze none.
Now you know everything you need to start running website experience surveys. What you should do next:
- choose the right website survey tool
- set up a survey using the template provided above
- analyze results keeping in mind described catches, especially browsers and devices
- implement changes on the website based on analysis
- create a new survey to see how changes improved the website experience
That’s it – you don’t need anything else. Just let me stress the importance of the 4th point of this framework. You collect and analyze data to improve the experience and it’s impossible without implementing changes on the website and running tests. With a tool like Optimizely, it’s not rocket science and anyone can do it.
And here’s the final tip: if you decide to run tests with Optimizely, you can prepare a few variations of the website. Using Survicate integration, you can target survey widgets at a specific variant so you clearly see how answers differ between variants. Learn more about this possibility here: https://survicate.construction/installation-guides/optimizely-integration/.