UI Evolution — How Tishi went from being disliked to loved
Recently, I started a new project in an unfamiliar territory — a mobile app to help women discover boutique jewelry designers from around the world. Needless to say, I’m not exactly the target group, and apart from buying a piece of jewelry here and there, I can’t say I have a lot of experience in this field. I stumbled into it after an intriguing discussion with my girlfriend, some friends and then a series of customer interviews.
The fascinating thing about jewelry is how often it is perceived as a luxury item for females, yet most traditional jewelry brands actually target men to buy pieces as tokens of love and appreciation for the women in their life. And after a 100 day tour through Southeast Asia, it became clearly apparent that jewelry pieces are much more than just luxurious accessories. It cuts across all demographics, from newborns to the elderly, irrespective of how wealthy you are, or not at all, permeates across all cultures, and is as old as mankind itself. Women especially, wear jewelry for a myriad of reasons — to highlight their features, express what they stand for, and connect with items that truly reflect their individual personalities, amongst many others.
When I built the first version of Tishi, I thought a “Tinder for Jewelry” would solve the problem women with a busy lifestyle experience — they want beautiful, unique pieces of jewelry, but lack time finding them. As it turns out though, the first version of the app was actually very much disliked. From there, we have iterated to something that is widely loved by our users now and is about to launch on the Apple App Store. Here is how.
v0.5: Colors, fonts & copy
Tishi had its logo designed whilst I was still playing around with the backend and picking up iOS development with Swift, as I previously only knew how to build Android apps with Java. The intention was to find a logo that resonated strongly with the target group and use it as the foundation for the app design, including colors and typography. Through constant testing of 113 different logos with a group of advisors and potential users (a process I might explain in another blog post if anyone is interested), we arrived at a dark purple color and serif font. However the font was not available on iOS and I wanted to keep the implementation simple for now, so we decided to test a few different fonts (in a similar process as we did with the logo) and found a (temporary) winner: Palatino (and later Georgia + Helvetica).
Since copy is not my greatest strength, I asked both my girlfriend and a copywriter for some input for the loading screen and key terms in the interface (e.g. should it just say “loading” or does Tishi have rotating tips and quotes? Should the wishlist label be “wishlist” or “tishibox”? Should the main screen be called “discover” or “explore”?) — all key parts of the user experience as I learned the hard way from previous projects.
v0.6: Functionality & UX
The above version was technically functional enough for a few beta testers to test the app (swipe right to save an item to your wishlist), but even though users liked the individual aspects I tested before (including the concept), the feedback once they played with the app was pretty devastating. At no point I thought about giving up, but instead of launching the app on the app store right away and leaving it with bad ratings for months until it’s at a decent state, I decided to skip the overheads of a live app and go back to the drawing board.
The most important piece of feedback was that just swiping isn’t enough — it wasn’t enough functionality to keep users excited and interested. Tishi needed a way for users to explore the whole collection in other ways. So I tested different concepts.
This step was done through quick mockups and ended up with us getting lots of feedback on which patterns people love, due to the usage of other apps. I also started changing our approach what kind of jewelry Tishi shows users. Instead of just sourcing items from shops around the web, we started working directly with boutique jewelry designers who sell quality pieces and love to share their own story and the stories behind their pieces, which our users are very interested in.
0.7: Navigation
Knowing that Tishi had to support different ways to explore jewelry, it became clear that the simple navigation of our first app was not enough. At the same time, a “sandwich” style menu to pack (and hide) functionality in seemed like a bad idea, as the UI would be strongly biased towards only one way of browsing the jewelry. After lots of testing, we decided to put the navigation at the top, so the swiping cards don’t overlap with the tab navigation.
If you are a one man band or a team consisting of developers only, it is worth noting that the process of finding the right navigation and doing the fine tuning was facilitated by an amazing UX designer whom I sought and paid. The impact can not be underestimated. Talking to and testing with customers gets you a long way to realise the problems, but finding the right solution is not always intuitive. While Apple has design guidelines, they are far from making sure you end up with something users would love.
0.8: Fine tuning.
With a designer on board, we did copious amounts of fine tuning on colours, layout, typography, flow and general user experience. I spent 2 weeks just tweaking little things in the UI to get them closer to the designs we had in front of us. Some items are still on the backlog because the delay they would cause does not justify their value right now, but all the little things we did added up to make a massive impact to the look and feel of the app.
0.9: Beta 2
No app is ever completed and no amount of up-front feedback can determine how successful an app will be; only the amount of traction generated from people actually using it. In this spirit, we decided to put Tishi through another beta and have started to gather another round of feedback for improvement. Most importantly, our Net Promoter Score has greatly increased and the app is ready to launch on the App Store soon.
I hope this blog post will encourage other developers to keep iterating on their app ideas, and work in conjunction with designers to turn a good idea into something beautiful.
If you have had any similar experiences or have tips to improve this process, I’d love to hear about them in the comments. If you happen to love jewelry, get Tishi here.
PS: I regularly share my lessons learned from building, managing and growing products at www.productrise.com. Also, if you are stuck with your product, team or personal Product Management journey, reach out to me any time. I’m always happy to just have a chat.
