Tailoring the Home tab to suit individual preference

User Research・UI Design

Makuake is a leading crowdfunding platform in Japan. I redesigned the Home Tab to help users find attractive items based on their interests.

Overview

With the rapid growth after our IPO, it became increasingly challenging for new users to find items of their interest on the Home Tab. This was primarily due to the lack of machine learning engineers available to develop recommendation features based on purchasing or browsing history, as well as insufficient human resources to curate a list of featured items.To address this, I devised a solution by introducing a new feature that allows users to request their favorite categories connected to tags. By selecting specific tags on the Home Tab, our system displays personalized recommendations based on the chosen tags. This approach enables us to provide relevant and tailored suggestions to users, enhancing their overall browsing and shopping experience on the platform.

Role

Design Owner (Full-time)

Deliverable

Mobile App

Timeline

May, 2020 - June, 2020 (2 months)

Team

Product Manager, iOS Engineers, Android Engineers

Final Design
Discover 🔍
I led user research by asking 8 users to demonstrate how they typically use the Makuake app.During the research, I identified a problem related to the items displayed on the Home Tab. Some users found it difficult to find items they were interested in on the Home Tab, as the list of products was ordered by the new arrival. Users' impressions varied depending on when they opened the app.
I didn't know I could find camping equipment here. How could I explore more?
- Male, 20’s
I can find only gadgets, and nothing seems to be my favourite.
- Female, 30’s
(Finding an item of women's underwear) I think I’m not a user of this item.
- Male, 50’s
Define 🎯
Since the release of the Makuake app, the Home Tab has displayed items in order of new arrival. However, with the growth of the platform, we were seeing approximately 50 new items added daily, which was increasing steadily. This resulted in the Home Tab displaying items updated randomly and frequently, which was not a good experience for users. Users had to rely on luck to find items they were interested in.
To solve this problem, I decided to personalize the Home Tab, so each user could find attractive items soon after opening the app.
Explore 💡

A: Recommend items based on purchasing or browsing history

Pros
If the accuracy of recommendations is high enough, it could increase conversion rates.
Cons
We have nothing to show new users with no history.
Constraints
We didn’t have enough machine learning engineers to develop recommendation features.

B. Create a list of featured items by each popular category

Pros
We can intentionally choose and promote items we want to sell more.
Cons
Users might get an impression of regular e-commerce. We wanted to avoid this as it is much different from the concept of crowdfunding, that new things are born.
Constraints
We need human resources to update the featured items. Users may get bored if there are less frequent updates.

C: Display items based on the tags users picked in order of new arrival

Pros
The list of items would be updated automatically without an additional workforce. And it suits the concept of crowdfunding: that new things are born from here.
Cons
If the timing of asking users about their favorite tags is not proper, it may annoy users exploring items.

Though Ideas A and B might make the Home Tab with an appealing list of items, they require a lot of human resources. Idea 3, utilizing tags, was a great solution to help users find items based on their interests from various perspectives. Moreover, it did not need additional human resources, which means we can try it immediately. We adopted this idea C. We thought asking users about their favorite tags during the onboarding experience would be good since it was mandatory not to interrupt users while exploring items.

Iteration

Iteration 1: Select an item, then follow a tag.

I thought it would be an opportunity to let people know about our popular items if it worked. However, during user testing, it turned out that people chose items based on their impression of the thumbnails, not their interests. Additionally, the tags connected to items were not always what they had expected.

Iteration 2: Select a category, then follow a tag.

Although this idea should have worked, unfortunately, the categories had not been updated since the service's release 6 years ago. Some categories had 30 tags, and the way they were divided was rough and not user-friendly.

Iteration 3: Follow a tag

Outcome
  • Time spent increased 1.5 x (60 sec → 90 sec)
  • The number of viewed items increased 1.5 x (4 items → 6 items)
Takeaways  💫
Even with constraints, effective solutions can be reached.
I faced a challenging situation where a recommended feature couldn't be developed due to constraints. Although disappointed, I didn't give up and instead focused on understanding the true problem that users were facing. I brainstormed and discussed ideas with a cross-functional team, and we came up with an alternative solution that turned out to be one of the most significant impacts I made on the company. This experience taught me that effective solutions can still be found even when faced with constraints. We can discover unexpected and innovative solutions that not only overcome constraints but also provide valuable insights for future problem-solving.
The user experience needs to keep updated.
Displaying items in order of new arrival helped Makuake establish its identity and achieve an IPO. However, user experience changes over time with product growth and environmental factors. Failure to update the product could lead to a decline in user experience and result in loss of users. Regularly engaging with users to identify current issues is essential to ensure the product is up-to-date and meets their needs.

Let’s get in touch

Feel free to reach out for collaboration or just say hello 👋

white mail icon