Introduction:
First and foremost, I want to emphasize that I am a seasoned developer with a decade of experience, born and raised in South Korea. In this article, I'd like to share my humble success story, taking into account that the web service I'm introducing is primarily known within South Korea.
Identifying the Gap in the Market:
Around 2019, MBTI-based psychological tests started gaining significant popularity in South Korea. Even today, online psychological tests continue to capture the interest of users. Capitalizing on this trend, several companies began offering SaaS services for creating psychological tests. However, these services were not only paid but also posed challenges in terms of content creation and complexity. Inspired by the idea that people find joy in taking tests and receiving results, I decided to create a service that allows users to easily create, solve, and share quizzes without the need for registration.
Development and Launch:
I rapidly developed the core functionality of creating, solving, and sharing quizzes in just one week.
Initial Marketing Strategy:
Once the website was developed, I faced the challenge of how to promote it. Without a specific strategy, I posted about it on internet community forums. Surprisingly, the response was overwhelmingly positive. Within the first month of launching, the website attracted 1.3 million users. This experience taught me that if the product is good, extensive marketing efforts may not be necessary. Engaging and interesting content tends to promote itself as users naturally share it.
The Turning Point:
Initially, I anticipated that users would create and share quizzes on various topics such as K-Pop, BTS, science, and culture. However, in reality, users started creating quizzes about themselves. These self-created quizzes gained immense popularity among elementary and middle school students in South Korea. Analyzing user patterns, I realized that users had a desire to express themselves and share their identities with others. Building on this insight, I added features such as an anonymous question-and-answer service and a snowman-building service. The snowman-building service, in particular, gained significant popularity in South Korea, attracting 1.2 million users in the past month.
Realtime (Universal Analytics)
Technical Challenges:
As is commonly known, the most significant challenge was the server load caused by a surge in users. To this day, I continue to operate the service with just a single physical server.
Tech Stack Used:
NGINX, Next.js, MySQL, Redis, Elasticsearch, Kibana, PM2
Handling Heavy Traffic:
To mitigate the server load resulting from influencers creating and sharing quizzes, I introduced caching, implemented database partitioning, and optimized indexing.
OG Image Generation:
To address the issue of generating thumbnail images for quizzes or snowman content to be shared externally, I switched to using Sharp. This change significantly reduced server load. While thumbnail generation with Puppeteer took approximately 1500ms, Sharp reduced it to about 100ms. Since switching to Sharp, I haven't experienced server load issues due to thumbnail generation.
The Next Steps:
Inspired by my modest success in South Korea, I became curious about whether I could achieve similar results in the United States. Although I have lived in Korea my entire life and don't fully understand American culture, I wanted to explore the possibility of my service succeeding in the United States. I may not know how to effectively promote it in the United States, but as a starting point, I decided to write this article and introduce my service experience.
Conclusion:
Here are the links to my services:
https://us.quizby.me (Quiz)
https://us.quizby.me/s (Snowman)
https://quizby.me (Korean version)
If you notice any grammar errors on the website or have suggestions for reaching American users, I would greatly appreciate your feedback. You are also welcome to create a snowman and share them with your friends.
Thank you for reading.