📡 Idea Radar 每天 8:00 / 18:00 推送 · 18:00 生成精选
reddit · SideProject · 48 分 ← 返回列表

I built something that tries to understand you from 5 things you love

I’ve been working on a little side project called Undertone . The idea came from something I’ve always found fascinating: the things we love might reveal more about us than we realise. Not just what we like, but the patterns in our choices. Undertone takes five things you love (films, TV shows, songs or books), asks a few questions about your preferences, and tries to find a common thread running through them. The result is a short, personal insight into what might be drawing

五维评分

付费意愿
40
需求真实度
30
个人可落地
90
竞争空间
35
新鲜度
55

AI 分析

痛点:用户愿意为“低门槛、一次性付费、无需注册”的自我洞察/趣味解读类小体验买单,而不是为明确功能痛点付费。

目标用户:喜欢电影/剧集/音乐/书籍、对自我认知和趣味测试感兴趣的年轻泛娱乐用户(英语市场为主)

证据:It’s a one-time 99p experience, with no account required. / I’ve tested it within a closed circle, and the responses have been encouraging. Now I’m curious whether it resonates with the rest of the world. / Would love to know whether the concept itself makes you curious enough to try it.

变现:一次性 99p–2.99 美元的小额付费解锁(无账号、无订阅);可延伸为分享卡片/深度报告加价、礼物卡、与流媒体或书影音平台的联名导流分成。

MVP:两周内可做:一个单页 Web 应用,用户输入 5 个喜欢的作品 + 几个偏好问题,调用 LLM 生成一段 200–300 字的个性化“共同线索”解读,先用 Stripe 一次性付款(首单免费额度换分享)验证转化率与分享率。

竞品:16Personalities 等性格测试类产品、Spotify Wrapped / 各类年度总结、GPT 套壳的 AI 命理/心理测试小应用、Goodreads、Letterboxd 等书影音偏好分析功能

摘要:作者发布了一个副业项目 Undertone:输入 5 个喜欢的电影/剧集/歌曲/书籍,AI 找出其中的共同线索,输出一段个性化自我洞察,定价一次性 99p、无需注册。作者称在小范围测试中反馈“encouraging”,但帖子没有任何真实付费或收入数据,需求验证仅停留在自述层面。独立开发者复刻成本极低(单页 + LLM API),但也意味着同类趣味测试产品竞争激烈、付费转化和复购是最大不确定性。