I worked with our Data Scientist to collect user performance for the last six months (2021/01/01 - 2021/06/30) on the mentor list page. And I found some interesting data that has room to grow.

Based on the data results, 5.6% of total users converted to sofasoda members, and only 0.71% of total users converted to purchase mentorship. Therefore, I started to think about how to grow the conversion rate? What's the "aha moment" that users will convert to sign up and purchase?

The filter feature wasn't as helpful as we thought because only 18.5% of users who use the filter clicked on our Mentor Profile page. In addition, the average view time that users stay on the profile pages before purchasing is 13.44 minutes.

Thus, the filter result might not be accurate for users, so they spent more time browsing mentors before purchasing.

After finding some data insights, the team interviewed our users to get more precise information. Then, we collected some critical feedback for this service.

After research, I hosted a brainstorming workshop and invited all the stakeholders to share their thoughts. In the workshop, we went through the data and interview insights. Then, we started brainstorming with four steps: define the problems, vote for problems, collect possible solutions, and vote for solutions.

Based on the research, the average view time that users stay on the mentor profile pages before purchasing is 13.44 minutes. As you can see, our users are struggling to find the right mentor for consultation. How can we help them make decisions?

We found our users most likely using their smartphones to browse the mentors. However, when purchased, they often change to using laptops.

According to the research, only 16.9% of users have used the filter. However, when we looked into the data, people who used filters saved a lot of time before finding the right mentor. How can we increase the usage of this feature to help our users?

It seems like the mentor description doesn't help users understand mentors. Considering having more efficient information on the card, like skills and highlights.

Over 50% of users browse more than three mentors before purchasing, which means our users think carefully. Therefore, having the save mentor feature can easily help them compare different mentors simultaneously.

We believe a more intuitive filter experience can help users save time looking for the right mentor. Thus, we want to increase user stickiness to the filter feature.

We compared the six-month data performance between the previous version (2021/01/01 - 2021/06/30) and the new versions (2021/07/01 - 2021/12/31). As you can see, both the sign-up rate and purchase rate increased (image 1).

According to the data performance, the sign-up rate increased for two reasons: the save mentor feature and campaigns. The first reason is users need to sign up if they want to save mentors to their list. Secondly, we hosted a free consultation lottery campaign in August 2021. During that time, users need to sign up to draw. Because of the campaign, the number of registered users has significantly increased. However, the event also decreased the percentage of filter usage (image 2).

To find out why the purchase rate increased, we looked carefully into different metrics. Firstly, our users are more likely to click on the result after filtering (an increase of 10.2%). It means the accuracy of filtering and the attractiveness of cards is much higher than in the previous version. Then, the time duration of page view before purchasing is a considerable improvement (a decrease of 7.52 minutes). Therefore, the new version helps users look for the right mentor faster and easier.