Weekly Dev Log #3


Happy Mid-Autumn Festival 🌕
Today is traditional Chinese festival(https://en.wikipedia.org/wiki/Mid-Autumn_Festival), Chineses people often gather together with their families on this day because it is a holiday for reunion.
I’m spending time with my friends. Being with friends is always so enjoyable.
I hope you can also spend a happy day with your family or friends.
About Dev
I've finally started restructuring LaxtTime's data model. Well, after I began, it wasn't as terrifying as I had imagined.
But I have to say restructuring really spent a lot of my time on design. I think it's always right to design a good plan before actually starting to write code.
The progress of restructuring is about 60%, I think this week I could publish a TestFlight version of LaxtTime 💪.
Bad News
I got some bad news: OneThing was rejected during the App Store review. The reason for rejection is 4.3(a) spam. Well, this is my first time getting a 4.3(a) rejection, and it doesn't feel good ☹️
But I really love my app, and I already have about 100 users. So I don't want to give up on it.
After completing the restructuring of LaxtTime, I will spend some time redesigning OneThing. This will be a thorough update, and I might even change the app's name.
ASO
In my previous logs, I expressed concerns about LaxtTime's distribution. 95% of downloads come from the Asia-Pacific region, while I've made significant efforts towards internationalization, such as setting English as the primary language and focusing on the needs of English-speaking users.
If the download situation continues like this, my efforts will not have yielded results.
This week, I made some ASO (App Store Optimization) attempts, and now I'm getting some downloads from outside the Asia-Pacific region.
Changed the subtitle. Now it's "Recurring reminder", before it was "Remember Last, Reminder Next"
I really loved the previous subtitle, but it was hard for new users to understand LaxtTime's function. I changed it to "Recurring reminder".
I think the vast majority of new downloads can be attributed to this change.
More keywords
I hadn't thoroughly researched keywords before, I just changed to a set of keywords that were more relevant to the app. Using an ASO tool, I found that my app's keyword coverage is still very low, only about 600, while similar competing apps usually have over 3000. This means they're more likely to be found in searches.
I will spend more time researching keywords. If it's effective, I'll share the results on my social media.
Ran an A/B test
This was my first time creating an A/B test for LaxtTime. I only changed one new preview image, but the new version actually performed worse. That's okay though, as I've learned how to create A/B tests. I like the saying "Don't guess it, test it", so I will create more A/B tests for ASO in the future.
Subscribe to my newsletter
Read articles from Hamming directly inside your inbox. Subscribe to the newsletter, and don't miss out.
Written by

Hamming
Hamming
indie hacker | Building something interesting I quit my full-time job to try building my cash flow through multiple products, in pursuit of true freedom. I'll document my independent development journey here and share everything with you. My product: iOS App: www.LaxtTime.top — $1.2k MRR