Small changes can have BIG results!
👋 Intro
Product design isn’t always about big glamorous new features — a little bit of user research and usability testing can open up an Aladdin cave of small, cost effective, opportunities! This is an example of just one small change we recently made to the Student Beans website which improved the way our users found relevant discounts - increasing our code usage by 65%!
👀 Overview
When our squad was tasked with improving our conversion rate, we made the decision to look at existing features, rather than creating new ones. We wanted to see if there were any areas of our site that could provide us with some low effort tweaks, but have a high impact on our conversion rate.
We identified our search as an area that could potentially have the biggest impact. It’s the single most interacted feature on our website and hadn’t had any TLC for over a year. We started to investigate how we could improve it based upon our users existing behaviours and also their expectations.
🔬 User Testing : How, why & what
We set up in-person and remote testing sessions to capture how our users interact with the search; when they used it, what for, and issues they had with it.
Our in-person user testing sessions usually consist of up to 5 testers. We ask them to run through tasks as two designers from our team alternate between taking notes and running the session.
The best results come from testing no more than 5 users and running as many small tests as you can afford.
Prior to the test we ensure that we have a clear agenda, set of tasks and goals we want to achieve. This is to keep us focused and guarantees a solid, workable set of results.
Our tasks for this session were search focused, for each task we make sure:
- They don’t lead the user into performing specific actions or using certain features. This means they are free to use the site as they would naturally and we get the most genuine observation.
- It is goal focused. Even if the user doesn’t achieve the desired goal, this is still a useful result as we can evaluate and work out why.
Once the sessions have been completed, we take away the notes and recordings and compile them into a report that is shared with the rest of the squad and sometimes the wider business.
🕵 Test Results: What we found
We found a flaw with what we thought was a pretty straight forward user flow from searching for a brand to getting a discount for that brand. It turned out that landing on a brand page prior to the discount page caused some confusion and in some cases ended their flow completely.
- We noticed users became confused as to where to click next. Despite the discount card being identical to everywhere else on the site, this was clearly not the best way of presenting the information in order to encourage a click and get the user to where they need to be.
- Users were also finding themselves at a dead end. A place no user should ever end up. This was a result of displaying brands that didn’t currently have an active discount.
We also found that although the majority of users were searching for a brand, they were looking for the discounts these brands had to offer. Which meant we had to be a step ahead of them in order to improve their experience.
🤔 The Idea
We could reduce a step in the journey and eliminate the page of confusion!
Our current user journey for search is:
- User searches for a brand (98% of search terms)
- look for a specific discount for this brand on the brand page
- Click on offer to redeem that discount.
The idea was to combine step one and two, in order to reduce steps and dead ends. We wanted to surface the discount at a higher level. We know that users search for brands first and foremost, so we were conscious that we needed to keep the brand name the most prominent piece of information in the results.
Our concern with this approach was if a brand had multiple discounts in the results. Would a user only click the first result and not notice the other available discounts from that brand? We decided as a squad to run this as a test and if it proved to be successful we would address this UX hole in the next iteration. Keeping experiments and iterations small!
⭐️ TL;DR — Our main areas to focus on would be:
- Stop showing brands with no discounts to remove dead ends
- Reduce the journey from search to discount redemption
- Show the search content in an easy to digest format
👨💻 Development Ready
We needed to test our assumptions as cheaply as possible so we made sure to use our existing components. We didn’t over think the UI here. If the experiments proved to be successful we can always push for pixel perfection then, if it doesn’t, we haven’t invested too much time and effort before knowing so.
📈 Results
We tracked the conversion rate for the same amount of time before and after we swapped brands to discounts (the small change) and the results showed an increase in conversion by 65% (the big result).
🚶♂️Next Steps: How can we take this further?
We will be performing more rounds of user testing on the search to see if the changes we have put in place are working as desired and are producing the results we hoped for.
We’ll also be looking into solutions for the issues and concerns that we found in the initial user testing:
- If we surface the discounts will users only click the first and miss any others in the list?
As a solution to this, we could look at adding the other available discounts to the bottom of the discount page. - Users not realising they need to click the discount card when on the brand page
We could add a CTA to be clear about the action the user needs to take in order to continue successfully with their journey - Dead ends generated from search results
We are exploring the idea of a larger search that would allow us to surface more results and information generated from different areas of the site, for example; recommended discounts, collections and categories.
