I've been obsessing over cold LinkedIn outreach for a while now. partly for my own pipeline, partly because I kept seeing people around me send completely reasonable messages and get nothing back.

So I started saving every opener that actually worked. Ones that turned into real conversations, not just a polite thumbs up or "thank you". Eventually I had about 50 of them across six industries.

Once I had enough I went back through all of them looking for patterns. Here's what almost every single one had in common:

1. Every message that worked ended with exactly one question. Not "thoughts on this? Also are you open to a call? I can also send a deck". Cutting the second ask is harder than it sounds but it's the difference between a message that gets answered and one that gets skimmed.

2. No pitch in the first message: Out of 50 messages, zero of them pitched anything in the opener. The only job of message one is to earn message two. That's it.

3. Signal-based beats evergreen every time: About 7 out of 10 of the messages that worked referenced something the other person did in the last week or two (a post, a hire, a product update, a comment they left somewhere.) The freshest openers consistently outperformed the generic reusable ones by a wide margin.

4. Specificity over cleverness: "Saw you hired three SDRs in Q3" will always outperform the wittiest line ever written. Not one of these 50 openers is clever in the LinkedIn influencer sense. Every single one references something real.

5. Peer framing, not target framing: The messages that converted treated the other person as an operator with an interesting problem, not as a budget to unlock. Ironically that framing tends to unlock more budget than any direct pitch ever could.

6. Always keep it short: Average length across all 50 is around 50 words. Nothing needed to be long to work. If a message needs three paragraphs to make its point, the point probably isn't sharp enough yet.

I ended up turning the whole collection into a proper doc organized by industry with a breakdown of why each one worked because the template is useless without understanding the reasoning behind it. The moment you paste one of these word for word to someone who didn't actually do the thing it references, it becomes spam like everything else in their inbox.

Happy to share the full doc in the comments if anyone wants it. It's free, no email wall or anything. just figured it's more useful out in the world than sitting in my notes.

Start from level 1 in The New Acquisition Game:

Client acquisition isn't just cold emails anymore.

It's now "Omnipresence + Automation".

Reframe getting clients as a game, and you'll become addicted to levelling up.

Level 1: Get Your First Wins

→ Launch a simple lead magnet that solves one problem
→ Send 50 personalized cold messages daily
→ Document your process in case studies
→ Build basic systems to handle inbound inquiries

Level 2: Build Visible Authority

→ Post your frameworks and lessons learned daily
→ Share client wins and transformations publicly
→ Teach what you know while building credibility
→ Give away valuable insights to prove expertise

Level 3: Scale Your Outbound

→ Use AI to research and personalize at volume
→ Test multiple channels: email, LinkedIn, video DMs
→ Create sequences that feel human, not robotic
→ Track what converts and double down

Level 4: Engineer Inbound

→ Create content that ranks and attracts your ICP
→ Build lead magnets that convert browsers to leads
→ Deploy nurture systems working while you sleep
→ Set up referral mechanisms that self-replicate

Level 5: Become Unavoidable

→ Your name appears when prospects research solutions
→ Content drives qualified leads without paid ads
→ Cold prospects already know who you are
→ Outbound becomes easier because of inbound authority

Level 6: Pure Inbound Mode

→ Prospects seek you out, not the other way around
→ Waitlists and application funnels filter clients
→ Community and partners feed you opportunities
→ Outbound is optional, not survival

Beat the game?

You're playing the long game now.

Early levels: Hustle meets strategy.
Final levels: Authority meets automation.

Anyway, If outbound is part of your work, this might be worth a look: verbatune.com

Volume + generic copy scaled fast because it felt like a cheat code. Everyone piled in. Reply rates tanked. The cheat code is now just the default, which means it's not a cheat code anymore, it's just noise.

Prospect research never got crowded because it takes real time. Checking what someone posted last week. Noticing they just hired three engineers or expanded into a new market. Finding the one line that makes the email feel like it was written specifically for that person and not pasted from a sequence template.

Nobody automated that part well enough to flood it. So it stayed open.

Shortcuts get you into a lot of inboxes. Research gets you a reply.

The funny thing is the research part is also the part that's actually gotten easier to automate recently, not the blasting part. Blasting a thousand emails was always easy. Finding the right signal for each person at scale was the bottleneck, and that bottleneck is mostly gone now if you're using the right setup.

The lane is still open. Most people are too busy blasting to notice.

Saw a stat somewhere that something like 60%+ of reps miss quota, and the more outreach I've done the more I believe it's not a copywriting problem. It's that everyone's working off stale data, no real process, and whatever free tool they found last week.

Personalizing with {first_name} and {job_title} isn't personalization, it's a find-and-replace. Without actual context behind it the message is still hollow, it just has a name in it now.

Here's the stuff I actually check before I write anything that gets me 20% to 40% reply rate:

1. Real signals like Funding round, new exec hire, event interest anything that just happened. this tells you WHY to reach out now, not just who to reach out to.
2. Actual intent, not assumed intent Did this person engage with anything relevant recently (a post, a topic, a piece of content) more than once? Someone who reacted to "AI sales strategies" four times this week is a very different conversation than someone who's never touched the topic.
3. Reference something real "Looks like you're into X, here's a resource on it" lands completely differently than a generic pitch, because it proves you actually looked before you typed.
4. Timing over guessing "Saw the new feature drop yesterday, that's a solid addition" works because it's tied to something that just happened. The same line sent two weeks later just sounds late.
5. Make it feel like a conversation not a script.
6. Be the one outreach that isn't generic Most outreach in anyone's inbox sounds identical. The bar to stand out is embarrassingly low right now, you basically just have to not sound like everyone else.

None of this requires being a stalker, it just requires actually looking before you hit send, sadly this is the part most people skip because it doesn't scale by hand.

I get it, the manual version of this gets old fast, checking signals and intent by hand for every single lead is a nightmare. For this, I ended up putting together my own setup that scans for the signals, pulls together the relevant info on each person, drafts the personalized message, and handles the follow-up sequence and booking on the back end, so the only part that's still actually me is showing up to the demo (if you automate it things will get so much easier).

Anyway curious what everyone else checks before sending, feel like there's always one more signal people swear by that I haven't tried yet.

I used to spend almost 10 hours per lead doing research, writing outreach, taking notes, writing proposals, and chasing follow-ups.

Now I spend about 5 min per lead. Same number of closed deals. (sometimes better)

The only part I still do myself is the actual discovery call, because human connection still matters there. Everything else is automated end to end. Here's the exact system (this is literally what my SaaS runs under the hood):

Step 1: Lead Generation (100% AI) → Scrapes LinkedIn + company databases → Qualifies leads against criteria (industry, size, tech stack) → Enriches contact data automatically → Delivers 250+ qualified leads a week, no manual list-building

Step 2: Outreach (100% AI) → Writes personalized DMs using real company research → Runs 3-5 touchpoint sequences → Books discovery calls straight to my calendar → 20-40% response rate

Step 3: Discovery Call (still human) → I run the call → AI transcribes in real time and pulls out pain points, budget, timeline, decision makers → Zero manual note-taking

Step 4: Proposal Generation (90% AI) → Takes the call transcript + my template → Generates a custom proposal: scope, pricing, timeline → I tweak it for 10-15 min → Sent within 2 hours of the call ending

Step 5: Follow-up (100% AI) → Sends the proposal with a personalized note → Tracks opens/engagement → Runs reminder sequences → Pings me the moment a prospect looks ready to close

I built this for myself first because I was drowning in busywork, not because I wanted to build a product. But after a few people in my network saw the workflow and asked "wait, can I use this," I packaged it into a SaaS. It's basically Steps 1, 2, 4 and 5 above, running on autopilot.

If anyone's curious how the lead scoring or the proposal-generation prompt chain works, happy to go into detail in the comments.