How to find customers on Hacker News (without getting flagged)
Hacker News readers are engineers, CTOs, and founders — people who choose and buy tools. But HN is also one of the most self-policing communities on the internet: obvious marketing gets flagged, downvoted, and remembered. Here's how to find real customers there while staying within community norms.
Where the buying signals live
1. Ask HN threads. "Ask HN: What do you use for error tracking?" is a purchase committee meeting happening in public. Search Ask HN [your category] on hn.algolia.com and sort by date.
2. Comment threads under competitor launches. When a competitor posts Show HN or gets covered, the comments fill with people describing what they actually need — including gaps the competitor doesn't cover.
3. Complaint threads. Pricing-change and acquisition announcements generate "time to migrate" comments. These are the warmest switching signals you'll find anywhere.
4. "Who is hiring" / "Freelancer" threads. A company hiring for a role your product replaces or augments is a signal.
5. Show HN posts from your ICP. A founder launching in your target segment just became a prospect for every tool a new company needs.
The HN Algolia search cheat sheet
https://hn.algolia.com/?query=%22alternative+to+X%22&dateRange=pastMonth— recent alternative-seekers- Search comments (not just stories) — most intent lives in comments
- Sort by date, not popularity; a 3-day-old comment with 2 points beats a famous thread from 2019
Outreach rules that keep you welcome
- Reply in-thread while the thread is alive (first 48h). A helpful, specific answer with disclosed affiliation ("I work on X, so biased, but...") is accepted and often upvoted.
- Never mass-reply. One thoughtful reply per relevant thread.
- If the thread is dead, go off-platform. Many HN users list emails or personal sites in their profile. Reference the exact comment: "Saw your comment on the Datadog pricing thread..."
- Don't fake grassroots. Voting rings and sockpuppet recommendations are detected and ban-worthy.
- Give more than you take. Answer questions where you have no product to sell; your comment history is your reputation.
Doing this at scale
Manually checking Algolia searches daily works but decays fast. LeadSleuth monitors Hacker News continuously against your ICP: it catches recommendation asks, competitor complaints, and problem descriptions, then builds a cited dossier on each poster (who they are, why the post signals intent, ICP fit) with a draft reply quoting their own words. You review and send; nothing is auto-posted.
More resources
Guides
- What are buying intent signals?
- Social listening for sales
- Using F5Bot for lead generation
- Reddit monitoring for sales
- Find B2B leads on Bluesky
- Find B2B leads on Reddit & HN
- Lead gen for devtools founders
- Lead gen for indie hackers
Comparisons
- LeadSleuth vs F5Bot
- LeadSleuth vs Syften
- LeadSleuth vs GummySearch
- LeadSleuth vs Trigify
- LeadSleuth vs Clay
- LeadSleuth vs Apollo
- AI SDR alternative
Also see pricing, the help center, or sign up free.
如何在 Hacker News 上找到客户(而不被 Flag)
Hacker News 的读者是工程师、CTO 和创始人——正是选型和采购工具的人。但 HN 也是互联网上自治最严格的社区之一:明显的营销会被 flag、被踩、被记住。以下是在遵守社区规范的前提下,在 HN 上找到真实客户的方法。
购买信号藏在哪里
1. Ask HN 帖。「Ask HN: 你们用什么做错误追踪?」就是一场公开进行的采购评审会。在 hn.algolia.com 搜索 Ask HN [你的品类] 并按日期排序。
2. 竞品发布帖的评论区。竞品发 Show HN 或被报道时,评论区会挤满描述自己真实需求的人——包括竞品没覆盖的缺口。
3. 吐槽帖。涨价公告、被收购公告下面全是「该迁移了」的评论。这是你能找到的最热的换用信号。
4. Who is hiring / 自由职业帖。一家公司在招聘你的产品所替代/增强的岗位,本身就是信号。
5. 目标客群的 Show HN。你的目标细分里有创始人刚发布产品——新公司需要的每一样工具,他都是潜在买家。
HN Algolia 搜索速查
https://hn.algolia.com/?query=%22alternative+to+X%22&dateRange=pastMonth——最近在找替代品的人- 记得搜评论(不只是帖子)——大部分意向在评论里
- 按日期而非热度排序;3 天前 2 分的评论比 2019 年的热帖有价值得多
让你持续受欢迎的触达规则
- 帖子还活着时(48 小时内)在原帖回复。有用、具体、披露利益相关(「我在做 X,立场有偏,但……」)的回答会被接受甚至被顶。
- 绝不批量回复。每个相关帖只留一条用心的回复。
- 帖子凉了就转到站外。很多 HN 用户在个人资料里留了邮箱或个人网站。引用具体评论:「看到你在 Datadog 涨价那个帖子里的评论……」
- 不要伪装群众。刷票和小号推荐会被检测并封号。
- 多给予少索取。在与你产品无关的问题下也认真回答;你的评论历史就是你的信誉。
规模化执行
每天手动查 Algolia 搜索可行,但很快会松懈。LeadSleuth 按你的 ICP 持续监控 Hacker News:捕捉求推荐、竞品抱怨和问题描述,为每位发帖人生成带引用来源的尽调档案(他是谁、为什么这条帖子代表购买意向、ICP 匹配度),并起草一条引用对方原话的回复。由你审核后发送,绝不自动发帖。