3
5 Comments

The cold email problem nobody talks about

Most cold emails fail before the prospect finishes the first sentence. Not because the offer is bad, but because the email is about the sender.

I spent 2025 sending 400+ cold emails a week. Zero calls booked. Not one. Looking back, every email had the same problem: it was a pitch disguised as outreach. "We help companies like yours..." No you don't, you help yourself.

Three things I've changed since:

The subject line test. If your subject line could be sent by any company in your space, it's generic. "Quick question" is the worst offender. The bar: would the recipient open this if they knew it was a sales email? If not, you're relying on deception, and that only works once.

The "so what" filter. Read every sentence and ask "so what?" after it. "We're an AI-powered platform that..." So what? "We've helped 50 companies increase..." So what, for me? If you can't answer "so what" for the reader within 2 sentences, cut everything above it.

Specificity over personalization. "I saw your LinkedIn post about X" isn't personalization anymore, it's a template with a variable. Real specificity: "You're running 120 listings on the Gulf Coast with a 62% direct booking rate, which means you're probably leaving $15-20k/month on the table from guests who visit your site and book on Airbnb instead." That takes 5 minutes of research. Most people won't do it. That's why it works.

I built a free tool that does this diagnostic automatically. Paste your cold email, it tears it apart line by line and tells you what to fix. No signup, no gate.

forgehouse.io/tools/cold-email-teardown

on March 9, 2026
  1. 1

    The "so what?" filter is genuinely one of the most useful editing techniques for cold outreach. Most people would fail this test on their first 3 sentences because they're leading with what the product IS rather than what problem it solves for THIS specific person.

    The specificity point is where I think most people get stuck. They understand the principle but 5 minutes of research per prospect is still 500 minutes for a 100-email campaign. What actually scales is getting the specificity into the segmentation itself — so you're not doing bespoke research per prospect but rather writing to a segment where that specific pain is structural.

    Your Gulf Coast vacation rental example is perfect: "you have 120 listings and a 62% direct booking rate" isn't just personalization, it's diagnosis. The prospect immediately knows you understand their business at a level most salespeople don't. But you can only do that if you've spent time deeply understanding that specific type of customer before writing a single email.

    What's been your experience with reply rates since making these changes? And are you running your campaigns manually or through tooling?

    1. 1

      You're right that segmentation is the real leverage. But I've found the 5 minutes per prospect is still the floor, not the ceiling. A GTM mentor I work with drilled that into me: if you can't spend 5 minutes understanding someone's business, you shouldn't be in their inbox. The segmentation just tells you where to look, the research tells you what to say.

      On tooling: I built a diagnostic that tears apart each email before I send it, basically a quality gate. But sending is still manual. Feels counterintuitive but I'd rather do 10 I'm sure about than automate 100 I'm guessing on.

  2. 1

    The 'so what' filter applies across every kind of email, but cold outreach and transactional email are solving completely different problems.

    Cold email: the reader doesn't know you exist and doesn't care about your offer yet. The fix is relevance — specificity that proves you did the work.

    Transactional email (payment recovery, invoice follow-up): the reader knows you, wants your product, and has a payment problem blocking them from continuing. The fix isn't personalization — it's clarity. 'Action needed — your payment didn't process' outperforms clever subject lines because the recipient is actively looking for a way to solve the problem.

    The failure mode in transactional email is the opposite of cold email: founders soften the message so much they bury the one thing the customer needs (a direct payment link). I've seen recovery emails with 3 paragraphs before the link appears. That kills conversions.

    Building that specific type of email sequence at tryrecoverkit.com taught me the other side of this coin — when you already have the relationship, directness is the personalization.

  3. 1

    I have just tried out and i love the output of this and also it's tell everything whats wrong or right.

    1. 1

      THank you very much! Glad it was useful. If anything felt off in the output, let me know. Still refining it.

Trending on Indie Hackers
Stop Building Features: Why 80% of Your Roadmap is a Waste of Time User Avatar 97 comments How to build a quick and dirty prototype to validate your idea User Avatar 50 comments Stop Spamming Reddit for MRR. It’s Killing Your Brand (You need Claude Code for BuildInPublic instead) User Avatar 41 comments The Quiet Positioning Trick Small Products Use to Beat Bigger Ones User Avatar 37 comments I got let go, spent 18 months building a productivity app, and now I'm taking it to Kickstarter User Avatar 29 comments I Thought AI Made Me Faster. My Metrics Disagreed. User Avatar 24 comments