The Problem

Many of Trunk Club’s customers were only making one purchase, many of these customers would purchase again but were unsure of how to get another trunk. If the stylist never contacted these customers again, most of them would never come back. We were hearing things like, “I had a great experience working with my stylist, but I am not sure how to get another trunk. Is there a way I can request one?” There was a clear need to provide our customers with a way to say when they wanted to hear from us and what they were interested in seeing.
 

Our Goal

We wanted to provide our customers with a clear way to request more trunks from us, tell us how often they wanted to receive them, and let their stylist know what they were looking for without making Trunk Club feel like a subscription service.

Discovery

Early customer satisfaction surveys helped show us that many customers expected a way to request and schedule trunks with their stylists. It was a feature that most of our key competitors had and relied on and many of our customers already had these expectations in their mind.

In order to hone in on what customers would expect from our execution, we held several research sessions—with TC customers and non-customers alike. Early sessions focused on asking questions about what people would expect to see if there was a way for them to request trunks. We found that most people expected to see some sort of way to select a date for when their trunk would arrive, and maybe a way to set up a frequency or cadence with their stylist. Many people also spoke of wanting a way to tell their stylist what they were looking for and why. 

Takeaways

We could see that needed ways within our application to let people schedule with us and tell us their style needs, so we started to make tactical decisions about how to get there. Our first step was to find a way to let people set a date for their next trunk immediately after transacting with us and leaving feedback

 

Designs

First Test

To validate the idea of scheduling with our customers, we rolled out a test to ~10% of our customers after they completed leaving feedback on a current trunk. The test simply asked if they wanted to schedule their next trunk and then showed them a calendar where they could select a week range for when they would like their next trunk. 

Since Trunk Club can't guarantee arrival dates, we decided to start by highlighting a full week range instead of a single date. We knew this wasn't ideal, but we wanted to at least start gathering feedback.

Initial Results

The initial test results were very helpful in showing us that a lot of customers were interested in scheduling with us. Of the people who saw the above test, close to half of them chose to schedule their next trunk

Recurring options and UI update

Now that we had a way to schedule the next trunk, we wanted to add in the ability to start a recurring schedule with us. While we were working to add this feature, we also updated the calendar UI a bit to reflect having a goal arrival date instead of highlighting a full week. 

We ended up having the scheduling page live within our Trunks page after sending out a card sorting survey that clearly pointed towards scheduling being tied to orders. 

One interesting thing to note is that we settled on the phrasing of 'Recurring schedule' after testing a few different phrases that didn't do as well. Initially, most of us thought the word 'Cadence' would work, but quite a few test participants were unfamiliar with the word.

Updating feedback modal

We also updated the modal coming from leaving feedback to ask how often people wanted to receive trunks from us.

Trunk Requests

Now that our customers were starting to get on a schedule (at a surprisingly high rate), our stylists were starting to run into issues of not knowing what to pack for their clients. When the time came to send a preview, there was very little for them to go off of. One of the other designers on our team (Jimmy Yoon) had been working on a similar project to capture customer needs that we had been calling ‘Trunk Requests’. We realized there was a significant overlap in the projects, and started to team with Jimmy to bring requests into scheduling. 

In Messenger

After most of these pieces were built and testing well, we wanted to give the feature a broader reach to our customers who mostly interact with Trunk Club via a chat with their stylist. We introduced a way to schedule and make requests from right within our messenger feature.

Outcome

Trunk Scheduling had a huge impact on our company. Customers who chose to get on a schedule were way more likely to transact with Trunk Club more frequently, and they often kept more items that were sent to them. Stylists were able to pack trunks and handle their book of clients more efficiently since they had to worry less about when to reach out and they had to spend less time asking about what items their clients were interested in seeing.

Ultimately, putting more control into our customer’s hands allowed for a better experience overall that better aligned with expectations of our service.
 

Reflection

This project was one of my favorites to work on at Trunk Club. Even though the final UI changes were relatively small, the amount of work that went into figuring out the logistics of adding scheduling to our business model and adjusting our process was incredibly challenging and rewarding. 

What went well

The rollout of this feature was relatively smooth, and our process of iterating in small chunks allowed us to monitor the impact of small changes as the feature evolved. The excitement about the feature from company leadership was very encouraging, and I felt like our team was making a huge impact

What I would change

Early on in this project, I was relatively new to the company and I was very guarded about my design process. As time went on, I became better about sharing early and often to engineers, other designers, and stake-holders. The more eyes I showed progress to, the better the final results were. If I could change one thing about how I approached this work it would be to have shared more early on.