📡 Idea Radar 每天 8:00 / 18:00 推送 · 18:00 生成精选
reddit · SaaS · 37 分 ← 返回列表

How do you fight the urge to build first instead lf doing marketing/product validation?

As a software engineer, I have this dopamine rush about building features for a new product despite knowing well the right thing to do is gather potential paying customers first then build the product. I always think I've done enough market research and then proceed to build a set of features. Then surprise surprise, I can't sell the product and I'm here pretending to be shocked... I always fall on this trap thinking if I build enough features, people will eventually come but it never wor

五维评分

付费意愿
25
需求真实度
30
个人可落地
55
竞争空间
35
新鲜度
60

AI 分析

痛点:代码中的 API、SDK、AI 模型弃用公告分散,开发者难以及时发现即将失效的依赖,可能导致产品故障。

目标用户:使用大量第三方 API/SDK/AI 模型的软件工程师、DevOps 和工程团队。

证据:I tried to build a waiting list first, then wait to see if people are interested before making something that works. If there is a single soul interested, the product I'm trying to build right now is called EOLTrace ... want to have proper validation from someone willing to pay for it. deprecations are scattered everywhere , not easy to find

变现:订阅制 SaaS:按仓库数、席位或 CI 扫描次数收费;GitHub App/CI 集成、Slack/邮件告警、私有化部署均可作为增值项。

MVP:2 周内做一个 GitHub Action:解析 package.json、requirements.txt、go.mod 等依赖文件,先覆盖 10-20 个高频 API/SDK/AI 模型厂商,用人工维护的弃用公告 JSON 比对,生成 PR 注释和邮件/Slack 告警;配一个 waitlist 落地页收集付费意向。

竞品:Dependabot、Renovate、Snyk、Endor Labs、GitHub Advanced Security、endoflife.date

摘要:Reddit 上一位软件工程师反思“先构建后营销”的陷阱,并提出 EOLTrace:通过本地扫描或 GitHub Actions 生成清单,追踪代码中的 API、SDK、AI 模型,并与厂商弃用公告数据库比对,提前发现即将失效项。当前仅在收集 waitlist,希望找到愿意付费的人验证需求,原帖没有真实收入或他人付费意愿证据。