July 23, 2020
2020 has been quite a year for the online revolution. Within weeks our grans were on Zoom, Generation-Z were doing school online and millions began working remotely.
An image doing the rounds on Twitter lately was asking people, who had led their digital transformation. Was it A. CEO B. CTO or C. COVID?
It’s almost as if we’ve fast-forwarded five years into the future. Which means businesses that want to survive can’t ignore going digital-first any longer.
But while digital transformation is one of those big scary phrases that everyone throws around without knowing what it means, getting started doesn’t have to be complicated.
All you need is a few days, some affordable tools and a handful of people.
How? With user testing.
Testing is at the heart of any great user experience and service innovation.
With the right strategy (which I’m going to share next) you can create a robust roadmap for improving digital products and services. One that is based on real scientific evidence.
Best of all, as COVID forced us to discover here at Tangent, you can do this remotely and cheaply.
Why Do User Testing?
There’s a famous saying that, “No plan survives first contact with the enemy”; well, in my experience, no plan survives first contact with your customer. If you put out a product without any testing at all, you’re going to make a lot of mistakes.
Likewise, without real evidence, it’s easy for decisions to be driven by your key stakeholders’ opinions.
And, if they’ve made the wrong assumptions, you’ll waste time and money on ideas that were always going to bomb – no matter how well you executed them.
The Exact 7 Steps to Follow
1. Start with the right test plan
First off, bring the key people together in a workshop to brainstorm and develop a test plan. You need to decide:
- What parts of your site, service or product you’re going to look at
- Which user personas or types of users you need to do the test
- What tasks or scenarios you’re going to test
- What your hypothesis is
- How you’ll create a prototype for the test
- What kind of interview you’re going to run
- What tool you’re going to use
To test remotely, we use LucidChart to put together the plan. Here’s a sample plan for you to use:
Fill out the form at the bottom of this page to receive your remote user testing PDF.
2. Recruit your panel of five
“Elaborate usability tests are a waste of resources. The best results come from testing no more than 5 users and running as many small tests as you can afford.” - Jakob Nielsen
In his seminal article, usability guru Jakob Nielsen found that a test with 5 users will help you find 84% of your problems.
Our own research bears this out. We’re always amazed at how people struggle in the same places.
This makes recruiting your test panel easier than you might have expected. Your best source is most likely your existing users or customers, who shouldn’t be difficult to recruit.
With a handful of people, you’ll quickly spot issues and opportunities and keeping the numbers down means you’ll be able to run more tests and gain more insight.
If for any reason you can’t use existing users, many of the tools we recommend will help you recruit a panel.
3. Pick your remote testing tool - here’s our review of the best
While traditionally we’ve run testing in-person, COVID forced us to rethink this entirely.
Over the past few months, we’ve tried and tested a number of different tools and found that they all work well in the right scenario.
While (spoiler alert) our winner turned out to be good old Zoom, the right testing tool for the job will always be the one which best suits your audience.
Here’s our roundup of the best ones out there at the moment:
Lookback.io offers the tech you need and they’ll also recruit a panel for you, which is a huge timesaver. If you’re launching a large B2C product, this would be our choice as it’s good value for money.
Userlytics is another useful all-in-one package that will help you recruit your panel and allow you to see and talk to your user while recording their screen.
If you want to do user testing on a larger scale, with both qualitative and quantitative insights, UserZoom will allow you to do it all in one place.
UserTesting is one of the most flexible tools we’ve tried as you can use it for mobile apps and real-world products and experiences. It also includes quantitative research methods for the best of both worlds.
PingPong is a good tool if you want to outsource and automate testing; from recruitment to transcribing tests. It also has a nice ‘emoji time stamp’ feature to help speed up the process.
Zoom has all the functionality you need to run a solid user testing programme, for free. With the ability to interview people while recording their faces, screens and voices, on both web and mobile, you can use Zoom to test practically any digital product.
4. Set up your test
A basic user test involves following along with your typical user as they use your service or product, asking them to ‘think aloud’ about how they’re thinking and feeling about the process; capturing what happens.
You’ll get the most accurate insights by creating the most realistic prototype as possible and most natural context. It also helps to get participants prepped in advance. If they’re not particularly tech-savvy or you’re using a platform they’re unfamiliar with, send them notes ahead of time or consider running a quick practice session. Remind them to make sure they have a reliable internet connection, headphones and a microphone so you can hear, see and record them clearly.
You’ll get the most accurate insights by creating the most realistic prototype as possible and most natural context. It also helps to get participants prepped in advance. If they’re not particularly tech-savvy or you’re using a platform they’re unfamiliar with, send them notes ahead of time or consider running a quick practice session. Remind them to make sure they have a reliable internet connection, headphones and a microphone so you can hear, see and record them clearly.
5. Listen, don’t lead, and take meticulous notes
Facilitating a good UX research interview isn’t a specialised discipline like design or development. It’s a skill that almost anyone can develop.
Essentially you’re just there to listen – analysis comes later. Most users are naturally good at voicing their thoughts aloud and won’t even need prompting.
In any case, it’s vital that you don’t lead them or you risk tainting the results.
Along with a facilitator, you also need someone else to keep notes with a timestamp log of what’s happening.
We use simple codes alongside each timestamp to indicate important behaviour such as where the user is struggling, where there’s confusion, system errors or hidden bugs.
However you decide to take notes, do it in a clear methodical way so you’ll be able to make sense of it more easily later.
6. Grab your post-its and analyse the results
A bit like TV thriller detectives pinning up vital clues on a whiteboard, you’ll need to mindmap your findings to see the bigger picture.
After reviewing your notes and time stamps, you create post-its of the main issues, stick them up on the wall then group them into themes.
We do this in teams, taking 30-minutes to discuss the findings of each test and agree where the big challenges are.
Working remotely - we do this via Airtable, where we can then turn them into individual line items to go through later.
7. Prioritise your roadmap
Once you know where your challenges and opportunities lie, you need to work out which will have the biggest return for your business and therefore what to focus on first.
We go through each theme and use the ‘PIE’ framework, to determine the priority – Potential, Importance, and Ease.
Here’s a quick rundown of how that works:
Potential We ask, how much improvement can we make with this change? If a page or product element performs poorly, then there’ll be more room for improvement.
Importance Next we ask, how important is this area of our product or site? If we fix this issue, how much of an impact will it have on our business? Highly trafficked pages are more important; there may be something that’s ripe for improvement but won’t make a great impact in the wider scheme of things.
Ease Finally we ask, how easy will it be to implement a change here? When we think of ‘ease’ we’re not just thinking about the technical side of things but factors such as internal politics and difficult sign-off procedures too.
To sum up? If a change scores highly for potential, importance and ease, then we know it’s definitely a high priority for the roadmap.
For us, testing is one of the most exciting parts of a digital design project. We learn tonnes and never fail to be surprised.
For example, when working with UK Power Networks we considered offering people who’d had a power cut a Just Eat or Deliveroo voucher so they wouldn’t be without hot food.
It’s the kind of concept everyone gets excited about in a meeting room but it only took 5 customers to tell us it was a terrible idea.
When we tested it with a prototype and told people they’d be without any power for another 3 hours, being offered the voucher actually made them feel MORE frustrated.
Had we not tested this, we would have invested huge amounts of time and money for a worse outcome.
Conclusion
Why test? Because the best insights don’t come from you, your design team or your boss – they come from your users.
The businesses that survive the COVID crisis will be those who stop guessing what their customers need and start listening instead.
I hope this article has shown that there’s no easier way to start doing that than with remote user testing. It’s also massively motivating for your team.
Within days you’ll have an evidence-based roadmap for simple changes that will make a big difference to your project.
Exactly the kind of plan that gives everyone new energy to crack on and improve things for the better. Which, after the year we’ve all had, is something we can all do with.
For more insights, tips and strategies like this check out my webinar Design Like A Scientist.
