What Happens to a Tactic the Day Someone Teaches It

I was on a call recently with a guy who has sent over a million emails. Not in his lifetime. I mean he tracks the number. A million-plus contact points, ten years in the business, selling high-value domain names - the kind of deals where one closed client can be worth more than most agencies make in a year. He's sold domains you'd recognize. He knows his market cold.

And he was getting his ass kicked by a competitor.

Not on product. Not on pricing. On infrastructure. On method. His competitor had a sending setup that looked - from the outside - like a custom SMTP running across a hundred-plus domains, spinning traffic through an exchange server that tied multiple domains together in a way that looked nothing like standard Google Workspace or Outlook sending. He'd gone through host.io, checked their DMARC records, dug into the DNS. They were clearly doing something different.

He wanted to know what it was. And he wanted to know why it was working when his setup - which was technically correct, using Instantly, 15 domains, three inboxes per domain - wasn't delivering results the way it used to.

The answer led us somewhere more interesting than infrastructure.

The Lifecycle of a Cold Email Tactic

Here's what I told him, and what I want to walk through with you, because it reframes every piece of outbound advice you'll ever read online.

My team has discovered methods, promoted them publicly, watched competitors steal them, watched them break - and then stopped promoting anything publicly ever again. That's not bitterness. That's pattern recognition. I've watched it happen enough times that it's basically a law now.

The lifecycle goes like this:

• Private discovery. Someone actually doing outbound at scale finds something that works. Maybe it's a sending architecture. Maybe it's a specific script structure. Maybe it's a way to combine two data sources nobody else has connected yet.
• Internal compounding. That person quietly uses it, refines it, scales it. Results are real. The edge is real.
• Public teaching. At some point - maybe because they want to build an audience, maybe because they're selling a course, maybe because they just want to share - they talk about it. Blog post, YouTube video, podcast, newsletter breakdown.
• Mass adoption. Everyone reads it. Everyone implements it. The tools that enable the tactic get overwhelmed with identical usage patterns. The email providers see it. The spam filters see it. The target inboxes see it.
• Death. The tactic stops working. The people who discovered it already moved on. Everyone who learned it from the blog post is left holding a dead playbook wondering what happened.

And the interval between each stage is getting shorter. What used to take two or three years to cycle through now takes months. Sometimes weeks.

So here's the implication that most people don't sit with long enough: the entire content ecosystem built around cold email is structurally incentivized to teach you what's already broken. The stuff that still works never gets published. The moment a tactic earns a write-up, it has already entered the dying phase. The people still winning aren't reading what you're reading.

The Google Workspace Example

The clearest version of this I can point to right now is Google Workspace sending.

A few years ago, the playbook was: get a Google Workspace account, set up your domain, warm it up, send 30 emails a day per inbox. That was it. That was the whole thing. And it worked. You could book meetings off that setup running basically naked - no spin syntax, no complex domain rotation, nothing fancy. Just clean emails sent from a warmed Google domain.

That playbook is everywhere now. It's in every free guide, every beginner YouTube video, every "cold email starter kit" lead magnet on the internet. And it's dead. Thirty emails a day out of a Google Workspace account in a cold campaign is not a strategy anymore. It's a way to burn domains.

Same with the three-to-four-users-per-domain setup. That works - technically. You won't get flagged immediately. But you'll get banned eventually, and when you do, you lose everything tied to that domain at once. It's a fragile architecture that's been widely taught, which means the providers have seen every variation of it and know exactly how to identify it.

The people who figured out those setups aren't using them anymore. They're using custom SMTP. And by the time custom SMTP is the standard free advice on Reddit, whoever's winning will be using something else.

What Custom SMTP Actually Is - and Why It Matters

The guy I was coaching had seen his competitor's setup and correctly identified it as custom SMTP. He wanted to understand how it worked.

The short version: instead of routing your email through Google or Microsoft's hosted infrastructure, you stand up your own mail server - typically on something like AWS - and send directly. You're passing email traffic the way you'd pass any other web traffic. Google and Outlook can't associate your sending domains together because they're not hosting any of them. There's no account to flag, no workspace to suspend, no single point of failure tied to a provider's terms of service.

Once you have your own server, you can spin up your own IPs. You can create your own proxy layers. You can generate as many IP addresses as you need. You're not renting infrastructure from someone who can pull the plug - you're running your own.

Our baseline is around 150 domains. We're sending maybe 14,000 to 15,000 emails a month off that, at roughly $300 total cost. That's not per domain. That's the whole operation. Three to four emails per domain per day, maximum. The safety is extreme. And because you're not on any provider's radar, your deliverability - assuming your content isn't garbage - lands in primary inbox, not promotions, not spam, and definitely not the shadow spam box that swallows the Nigerian prince emails and the Viagra pitches before they ever show up anywhere.

100 domains isn't an aggressive setup. 100 domains is table stakes. I wouldn't even call it a beginner configuration.

Now - is this complex to set up? Yes. I'm not going to pretend otherwise. This is why I have a head cold email coach who manages this specifically and sends over a million emails a month himself. When I told the guy I was coaching about this, his immediate question was whether there were resources to learn it. There are. But the honest answer is that some of this you're going to want to work through with someone who has already solved the problems, because the documentation is scattered and the failure modes are expensive.

Why Deliverability Beats Copy

Here's something that sounds counterintuitive until you think about it for two seconds: deliverability matters more than your script.

Not forever. Not in all contexts. But right now, in the current environment, getting the email to the inbox is the primary constraint. Everything else - your subject line, your opener, your call to action, your personalization strategy - all of that is downstream of whether the email actually arrives.

I told the guy I was coaching: it doesn't necessarily matter if you're using some famous copywriter's script or whatever framework is trending this month. If your email doesn't land in the inbox, none of it matters. The best-written cold email in the world sitting in a spam folder has a zero percent reply rate.

This is also why the advice to "just send better emails" is often the wrong lever to pull when campaigns stop working. The problem usually isn't the writing. The problem is that you're using an infrastructure setup that's been documented, tagged, and filtered - because someone taught it publicly and now every provider knows the fingerprint.

Fix the pipe first. Then optimize the message.

The Lead Problem Is the Same Problem

The same lifecycle that kills tactics also degrades data sources.

The guy I was coaching had a smart approach to finding buyers for high-value domains: go to Crunchbase, find companies at growth inflection points, look for the ones likely heading into a funding round, pull their domains, push them into Apollo to find decision-makers. Logical. Targeted. And genuinely useful for his specific business where you're not blasting a list - you're identifying a small, specific set of companies that would actually want to own a particular domain.

His complaint was that Apollo wasn't returning hits on a lot of those Crunchbase companies. My take: Crunchbase is better for certain slices of data - high-growth, venture-backed companies - but it doesn't capture legacy businesses, bootstrapped companies, or anything that doesn't fit the VC-trackable profile. Apollo has more raw volume. They say 240 million contacts. I don't trust that number entirely, but I'm confident in at least 50 to 60 million usable records, which means if you can master the search logic, you can build lists that Crunchbase-only searches will never surface.

The key is keyword search, not just industry filters. Apollo lets you search company keywords - the tags and descriptors embedded in company profiles. If you're selling a domain like longevity.com, you're not just searching "health" as an industry. You're searching "longevity" as a company keyword. You're iterating through variations. You're finding the specific pocket of companies that would actually care about owning that exact domain, not just a broad category that might be adjacent to it.

Master your Apollo search, feed those results into ScraperCity to pull and process the lead data, run it through an email verifier like NeverBounce, and you've got a sendable list. That's the whole loop. It's not complicated. The complication is in the keyword discovery - figuring out exactly what search terms surface the right companies. That's the part that requires actual domain knowledge about your market.

For finding verified email addresses specifically, ScraperCity's email finder is worth running your contacts through before you import into any campaign. Garbage in, garbage out - and sending to bad addresses kills deliverability faster than almost anything else.

The Enrichment Layer Most People Skip

One thing that came up in our conversation that I want to highlight because most people running cold email campaigns aren't doing this: lead enrichment after the initial scrape.

The basic flow everyone knows is: build list, verify emails, send campaign. That's fine. But if you want to layer in other channels - cold calls, direct mail, LinkedIn - you need more data than a name and an email address.

My current flow for this uses n8n (the automation platform, which I prefer because it's visual, self-hostable, and flat-rate once you run it on your own server). The workflow: pull raw leads from Apollo via ScraperCity, then run an n8n flow to enrich them. If I want physical addresses, I use a skip trace API - you're paying maybe $7 per thousand results, so fractions of a penny per record, and you get all the public-database-associated addresses for that person. Same search also surfaces personal phone numbers.

I actually use this for cold outreach via handwritten letters. There's a service called Scribless that has a literal robot writing letters by hand. You feed it an address and copy via webhook, it writes the letter, it mails it. Costs about 25 cents per piece. For high-value domain sales where one deal is worth tens or hundreds of thousands of dollars, spending 25 cents to get a physical piece of mail in front of a decision-maker is an absurd ROI calculation. That channel isn't saturated. That channel isn't getting blog posts written about it. That channel is working.

For the phone number side specifically: the skip trace approach is cleaner than most dedicated phone finding tools for consumer-adjacent searches. You search by name, or by name plus address if you have it, and then you run an AI verification step to match the returned phone numbers against the name - confirming that the number actually belongs to the person you're looking for before you dial. It's not perfect, but it's the closest thing to a modern phone book that exists.

If you want the list-building fundamentals laid out in a more structured way, the Best Lead Strategy Guide covers a lot of this ground.

The 30,000 Dead Leads Problem

He also had a list of 30,000 to 40,000 contacts accumulated over ten years. The problem: people change jobs. A contact that was a VP at a fast-growing startup five years ago might be at a completely different company now - or might be the founder of their own thing - or might have left the industry entirely.

Sending to a stale list is actively harmful, not just ineffective. Hard bounces from outdated emails hurt your domain reputation. Replies from people who are annoyed that you reached them at a job they left two years ago don't help your relationships.

The fix is a job change verification workflow. The one I use in n8n: loop through each contact in a Google Sheet, find their current LinkedIn profile, check whether they still work at the company you have on file. If they've moved on, the workflow finds whoever replaced them in that role - and then finds the new person's email via LinkedIn scraping. This way you're not just maintaining a list, you're actively keeping it pointed at the right people.

Important note on how to do this without torching your own LinkedIn: ScraperCity's LinkedIn scraper handles the actual profile pulls via its own infrastructure, not your personal account. Your LinkedIn never touches the scraping activity. This matters a lot if you've built up a LinkedIn presence with real followers and real connections - you don't want to risk that account for automation that can be done another way.

The People Still Winning Aren't Reading What You're Reading

I want to come back to the core idea here, because I think it's the most important thing in this post and I don't want it to get buried under the technical details.

The reason my competitor - and your competitor - seems to have a setup you can't quite reverse-engineer from publicly available information is not because they're smarter than you. It's because the thing they're using hasn't been publicly written about yet. By the time you read a detailed breakdown of it somewhere free on the internet, it'll be entering the dying phase. The mass adoption will have already started. The window will be closing.

This is why I stopped publicly promoting the methods that are currently working. We came up with our current approaches, promoted them early on, watched people steal them, watched them degrade. That's what taught us to keep the working stuff inside Galadon Gold and off the public internet. Not to hoard information. But because the moment something gets published, the clock starts. And I'm not in the business of teaching dead tactics.

The guy I was coaching had the right instinct. He saw his competitor's setup, tried to reverse-engineer it from public infrastructure signals, and came to the conclusion that they were doing something fundamentally different - not just a variation of what he already knew. He was right. And the reason he had to come to a paid coaching call to learn the specifics rather than finding them in a free guide is exactly the dynamic I'm describing.

Free content teaches what's already broken. It has to. The people still winning don't have an incentive to tell you what's working until it stops working for them - at which point they'll happily write the blog post, because it doesn't cost them anything anymore.

If you want the stuff that's currently working, you have two options: run enough volume yourself to discover it through iteration, or get inside a room with people who are actively doing it at scale. There's no third path.

If you want the scripts and frameworks that are at least current enough to be worth testing, you can grab the top 5 cold email scripts here - but read them understanding where they sit in the lifecycle. And if you're serious about getting your outbound infrastructure built correctly before the window on current methods closes, that's what Galadon Gold is for.

The call I described at the top of this post was 15 minutes. Fifteen minutes to map out an infrastructure upgrade, a lead sourcing strategy, an enrichment workflow, and a list hygiene process for someone who had already spent years figuring out the hard way what happens when you try to learn this stuff from public sources alone.

The information on the internet about cold email will always be one cycle behind. That's not a bug. It's the structure of how this ecosystem works. Build accordingly.