- User testing is the best form of research you can do
- Used in isolation it will uncover some conversion blockages. However, used as part of a research team it will uncover and validate many conversion blockages that can be fixed based upon priority
In the summer of 2015, I was asked “if you could use only one form of research other than heuristic analysis, what would it be and why”?
Instinctively I thought of Google Analytics (GA) because it provides information that is essential to all conversion optimisation projects. It provides so much useful information (when set up correctly) like conversions by device type, site search, high traffic, high bounce rate, etc. Fundamentally it tells you what is happening on your site which is powerful stuff.
But, the more I thought about the question, the clearer it became that actually I wouldn’t choose GA as my sole piece of research.
Instead I’d choose a qualitative piece of research. I’d choose User Testing.
User Testing for me is like Lionel Messi. It is the star player amongst a great team of research methods.
Google Analytics is the consistent player that provides useful information and tells you what is happening, but User Testing provides so much more in highlighting why things are happening on your site.
- Why are people bouncing from your site?
- Why are you losing a lot of visitors on your category pages?
- Why does your search function not convert the way you expect?
- Why are customers not proceeding to your checkout?
- Why are customers dropping out of your checkout?
All of these questions, and more, User Testing has the power to answer.
It gives companies the chance to get inside the mind of their visitors. It helps pinpoint friction points and helps to explain the causes of anxiety for existing and potential customers.
Once you know the causes and why something is or isn’t happening, you can then go about creating the solution that will potentially fix the problems your customers are experiencing (in priority order).
Another reason I’d choose User Testing is because it allows you to trial the fixes in the lab or remotely before you actually put your scientific experiment live to your sample size. This is very useful.
Is this how your customers feel when they checkout?
Here’s an example of remote User Testing carried out on a pet website, the participant had to buy food and complete the checkout process. The checkout process required customers to supply a lot of (unnecessary) information, would only allow them to buy if they spent over a certain amount, and they had to provide pet names and veterinary practice details. It was painful to watch them struggle with the website. No other single piece of research would provide the insights that shows how difficult the experience is. If you’d like to watch it here it is. It’s 11 minutes long and started to get really interesting after 4 minutes!
Going back to my football (soccer) analogy, here is my starting eleven research team (Research FC)
User test every time
So, back to the original question I was asked, “if you could use only one form of research other than heuristic analysis, what would it be and why?”
My answer would be User Testing. It would be my star player in the team. But it would have to be part of, and play for the benefit of, the overall team.
Whilst Google Analytics may be the holding midfielder who does a lot of great unassuming work and User Testing may be the striker who scores the goals, they can’t exist alone or as a pair, they need the other nine players.
An individual player can win you a game, but a team will win you the league.
Look at your research elements as a team and used together they will find where you are losing money on your site and play an important role in helping you get back that lost money that is rightfully yours.
So that’s my answer, I’d be interested to know what your’s would be?