Have you spoken to a member of our conversion optimisation team in the past few months? If you have, you’ll probably have heard about how much we love remote user testing.
Naturally, there are plenty of remote user testing tools out there for you to choose from. And, generally speaking, these enable you to recruit a panel of users, ask them to complete a set of tasks on your website or app and provide live feedback as they do so.
An example of one of our tests from Userlytics.com.
Unlike standard screen recordings or other ways of collecting qualitative UX data, the immediate nature of remote user testing allows you to listen to the user as they complete a task. This means you receive incredibly rich information about why a user has or has not taken an action.
5 Tips for Better Remote User Testing
Conversion being our line of business, we’ve run countless remote user tests. And it’s fair to say that we’ve picked up a thing or two along the way. So, without further ado, here are five easy and effective ways that you can get more from remote user testing.
1. Plan Your Task Using Google Analytics Data
Looking at user data through Google Analytics is a great way of understanding how people are currently using your website. Start by looking at the navigation summary of users on one of your top landing pages and subsequent pages to get a better idea of a typical user journey.
Conversion funnels can be used to pinpoint opportunities for improvements to site performance. And there are a few other starting points for user testing task planning:
- Find key user drop-off points using your conversion funnel
- Use your site content report to identify pages from which users are exiting your site
- On an e-commerce site, you can use the product performance report to identify products with low basket-to-detail rates
- Compare the bounce and conversion rates of users who land on different landing pages to identify under-performing content
Use this information to plan your task. If the person testing your website is encouraged through a similar process to a typical user, the feedback you receive will be more relevant and useful.
2. Clearly Outline Your Objectives
User testing is incredibly valuable because it gives you the ability to uncover issues that you wouldn’t otherwise have found. And, of course, the way to get the most out of your user testing is to outline the objectives of your session.
Objectives like "I want to uncover the reason that 35% of my users are dropping off throughout the checkout funnel" or "I want to ensure that there's nothing stopping my site making the best first impression on my users" can help drive effective user testing task creation.
In turn, a structured and goal-driven approach offers you the best chance of uncovering useful and actionable insights that will, ultimately, impact your key performance metrics.
3. Build a Panel Representative of Your Audience
Does your audience know everything about football? Maybe your website’s visitors are all finance experts or gardening enthusiasts. Regardless of your target audience, it’s essential to ensure that your testers have the knowledge required to use your site in the same way a typical user would.
A tester who doesn’t understand niche words on your site or understand what they would be expected to do on your site will not be able to offer you the same level of insight into the typical behaviour of your average user.
Screening questions are an effective way of ensuring you have the right user testers. For instance, in a recent user test for a client in the mortgage industry, we asked participants “have you previously, or are you in the process of applying for a mortgage?”.
This question meant we had ensured that our testers wouldn’t be negatively impacted in their in their ability to complete the task. If users are properly screened and still struggle to understand your site, then this can highlight a genuine issue which can’t be ignored on the basis that “a real user would understand this”.
4. Don’t Use Leading Questions
When you ask a tester “does this site look trustworthy?”, the vast majority will say “yes” and not much else. By using open questions like “what is your first impression of this site?” you will get a more genuine response that allows the tester to focus on what stands out to them on that page.
You can also be less specific, replacing questions like “do you know how you visit the store from this page?” with “what do you think you are expected to do here?”. Another example might be replacing “is there enough reassurance about this product’s quality for you to consider purchasing it?” with “is there anything that’s stopping you from buying this product today?”.
Leading questions make it harder to find new issues as they only help reinforce our opinions regarding previously uncovered issues. This reinforcement is often not genuine as the tester may feel prompted to give you the response that you are looking for - a result of response bias.
5. Ask Testers to Think Aloud
As marketers we naturally want to try and get our hands on as much data as possible. And when it comes to users testing this is as important as ever.
Ask your testers to think aloud and reassure them that there are no wrong answers - this means you get visibility of their entire thought process. Often, information that doesn't seem important to the user can actually be very useful for gaining key insights.
This information can make it a lot easier to identify issues that relate to your site’s navigation, user interface, reassurance messaging and design. If the user isn’t thinking aloud, it can be difficult to attain this type of information without asking leading questions.
Bonus Tip: Sample Sizes
While it is tempting to have as many users testing your site as possible, it has been found that it only requires five people to uncover the majority of issues on your website.
Asking more users than this to complete your tasks is likely not the most effective use of resources. Instead, it’s probably a better idea to direct resources towards having users complete different tasks or to complete them on different devices.
Also remember that not every response is going to be representative of each of your users.This is especially true when it comes to subjective aspects of your site; colours and images, for instance.
In a recent user test we had one user spend a significant amount of time telling us how much they liked the header image on the homepage, while another couldn't comprehend why this image could've possibly been chosen.
In cases such as this, we like to use Usabilityhub's five second test. This tool allows you to see what a number of users thought about your site in the first five seconds of seeing your homepage. This approach offers a much clearer picture of the first impression your website is making.
If you’re in need of further guidance regarding user testing, or would like a discussion about conversion optimisation generally, please don’t hesitate to get in touch here...
