Conduct Research for conversion opportunities [Part 2]

In the previous article, I talked about these ways to conduct a research to discover conversion opportunities:

• Technical analysis of the website dedicated for conversion
• Site Walkthrough
• Heuristic Analysis
• Usability evaluation

In this article, we’ll explore more ways, all complementary, to gather insights from your customers to understand them more and see what’s not working in your website.

We’ll explore:

• How to create surveys the right way
• User testing: The why and the how

(One other way will be reviewed in the future and that is how to make sense (gather insights) from Google Analytics)

Before anything, give me my candy

If you want to get responses to your surveys and if you want to get customers to be your user testing participants, you need to provide an incentive. Something like a chance to win a 100$ Amazon gift card would work. For user testing though, that’s a job, you need to pay them.

How to create surveys the right way

Surveys are a very great way to gather both qualitative and quantitative data. Typically in a survey, you may want to include both quantitative questions like how many times a day you check your baby diapers and qualitative questions like what’s the most challenging part of the postpartum period.

The people you want to survey are people who very recently did a purchase on your website. If you let time go by, well, you may end up with false information due to false memory (different from what actually happened). The minimum number of people you want to survey is atleast 100. Otherwise, your results won’t be statistically significant.

The goal of the survey when trying to optimize your website, is to understand what’s causing friction on your website. So questions like “What was the thing that almost held your from completing your purchase” are really valuable. You need to also uncover if there’s any question they didn’t have an answer for and what doubts were they having. Another important point to focus on is to uncover if they were comparison shopping. If they did, what brands they were exploring and what features or benefits they were comparing.

When it comes to the tools to be used, it doesn’t seem to be a big issue, there are a lot of them out there and most of them are enough. Google Docs, Typeform or whatever other solution are good enough so don’t try to overanalyze what’s the best.

User testing: The why and the how

The right way right? well, our goal here is to do our best and try to avoid some common mistakes and be aware of our own biases as well as the participants.

While analytics shows us what happens exactly, for example: If 10,000 people have landed on your product page from a PPC ad, 9000 added the product to their cart but only 500 proceeded to checkout, it’s crystal clear that there’s something wrong happening in the cart. Now you any uncover why this is happening with a heuristic evaluation (you going through to checkout and trying to see if there’s a usability problem, a friction issue, a technical issue probably), but sometimes the problem is in the your target audience own brains.

So to uncover that, you need to conduct interviews.

Here’s my process of conducting a user testing:

• Ice breaking so the environment will be less tense for the interviewee. Stress is a string emotion and it changes the user behavior, so we need to make sure the user is comfortable.
• Starting with questions about the whole problem. If it’s a yoga equipments ecommerce, well start with yoga. This is a good opportunity to understand the customer.
• A Slow (and hopefully they wouldn’t notice) transition to the website.
• Understand if they know what the website is about. The goal here is to evaluate the clarity of your messaging.
• Give the tasks to be done. Try specific tasks and see how they do. Will they be able to search and filter for a specific product? Is the checkout flow simple enough?

10 to 15 participants would be great for your user testing data.

User testing limitations

Having people to buy your product because they’re in a user testing session isn’t ideal obviously. The ideal would be to see real world users shopping on your website while having the ability to read their minds. We can’t do that and hopefully we never will. So you may want to use tools that record user sessions and then you can watch videos of the screen of your users when browsing your website or making a purchase. However this still isn’t ideal. It’s true that these are real customers doing the real deal thing but you can’t uncover why they existed on a certain page. All you’ll have is assumptions.

User testing is also as good as our abilities to understand humans. Sometimes they may tell us something they think it’s what we want to hear. So the limitation here is when they fool us. ALso, they may say something they think it’s true but when tested, it doesn’t make sense. For example, they probably would say, this popup is annoying and I want to exit the website because of it. However, when we test the popup, we find that it actually increases conversion.

For me, I’ll try to dig more into psychology more. That’d a great step to level up my user testing game. And I encourage you to do the same.

This field is hard. It really is, it requires good skills in copy, design, an understanding of how businesses work and how people actually work (psychology, neuroscience). So our duty as conversion optimizers (and good designers and copywriters) is to keep developing ourselves and learning about this field.

So this was another article that I write to share what I’ve learnt in the CRO (conversion optimization) learning journey.

This the sixth article of 12 articles I’m going to share what I’m learning from the CXL conversion optimization minidegree. One every week.

So if you want to join me learning, enroll in their in their minidegree or follow me if you need some more motivation. I’ll give you 6 more.