Creative Ideas Campaign: Cold Email Systems That Scale

Two Cold Email Videos Worth Your Time This Week

I watch a lot of cold email content. Most of it covers the same ground: write shorter emails, personalize the first line, follow up three times. Useful baseline stuff, but nothing that makes you stop and take notes.

These two videos stood out this week. Not because they're perfect, but because they each contain at least one framework that serious outbound operators should be thinking about. I watched them so you don't have to. Here's what's actually worth stealing, and where I'd push back.

Video 1: The Creative Ideas Campaign (Instantly Channel)

What the Video Is About

Eric Noslowski from Growth EngineX walks through a campaign framework he originally built for Instantly and has since deployed across dozens of clients. He calls it the Creative Ideas Campaign. The premise: instead of leading with a pitch, you lead with three specific ideas about how you could help the prospect grow their business. Each idea is AI-generated, based on that company's website content, target market, and positioning.

The email structure looks like this: a short opener acknowledging their business, three tailored campaign ideas, a brief value prop, and a CTA. He also shows a variation where you use one idea in the body and drop a second in the P.S., which keeps the email visually short while still delivering two angles.

The Data He Shows

This is the part that got my attention. Eric walks through a split test he ran for Instantly across multiple campaign types, showing contacts-per-opportunity for each:

• Newly hired team members in ICP: 176 contacts per opportunity
• Competitor platform users: 296 contacts per opportunity
• Founder sales automation: no meaningful results
• Case study matching: 341 contacts per opportunity
• Two-sentence direct email: 175 contacts per opportunity
• Technographic targeting: 238 contacts per opportunity
• Join a free community: 81 contacts per opportunity
• AI-generated creative ideas campaign: 85 contacts per opportunity

The free community number is interesting, but he dismisses it quickly because it's not monetizing directly. The creative ideas campaign sitting at 85 contacts per opportunity while being infinitely scalable is the real story here. Most of the other campaigns are capped by a trigger, a competitor list, a technographic filter. This one scales to any prospect you can find a valid email for.

My Reaction

The scalability argument is the strongest thing in this video. When you're running campaigns tied to triggers like job changes or competitor tool usage, you're always working against a shrinking pool. A newly hired VP of Sales is only a new hire for 90 days. A technographic filter depends on data that goes stale. The creative ideas approach removes that ceiling.

I've seen this play out firsthand. When I was building out outbound systems for clients early on, the campaigns that crushed over long periods weren't the ones tied to clever triggers. They were the ones where the angle stayed relevant to any company in the market, any day of the year. Good ideas don't expire.

The part I'd push back on is the length. Eric acknowledges the email is longer than what most cold email practitioners recommend, and he defends it by saying if the ideas are genuinely good, people will read. That's true. But it's also a risky bet if your AI prompt isn't dialed in. A mediocre three-bullet AI email is not going to get read. A sharp, specific one-line idea with a P.S. variation will outperform a bloated template almost every time in my experience.

His own P.S. format example for a hypothetical Red Bull campaign is better than his three-bullet format: "Hey, we could help you reach out to colleges and universities to make sure your product is stocked up for the students... P.S. If colleges isn't the focus right now, maybe we could help you target co-working spaces to see if founders would use it in the vending machines." That's sharp. Two ideas, short, punchy, and specific. I'd lean into that format over the three-bullet version.

The Prompt Engineering Setup

Eric walks through how to build the underlying AI prompt, and there are three things worth taking from it. First, define what ideas are off-limits. If you're a pharma expense management company, you don't want your AI generating ideas about accounts receivable. Constrain the output before it starts. Second, write the email manually for 10 target companies first. Use those as training examples inside the prompt so the AI knows the tone and format you want. Third, use cached inputs. If the beginning of your prompt stays identical and only the company-specific data changes at the bottom, OpenAI and Anthropic will charge you less because they've already processed that prefix. Small savings per email, significant savings at scale.

For building and deploying this kind of campaign, Clay is the tool he references for enriching company data before it hits your prompt. That makes sense. You need website content and basic firmographic data to generate ideas that don't sound generic. Without good input data, the AI output is going to be weak regardless of how clean your prompt is.

If you want pre-built scripts for campaigns like this, my Cold Email GPT Prompts pack covers the prompt structure in detail.

What to Implement vs. What to Skip

Implement: The one-idea-plus-P.S. format. It keeps the email short, delivers two hooks, and gives you data on which angle resonates. Implement: Writing the prompt manually for 10 companies first before automating. That training data step is what separates campaigns that generate interesting ideas from ones that generate generic noise. Implement: Caching the static portion of your prompt at the top. Free money.

Skip or test carefully: The three-bullet format at scale. It works if the ideas are exceptional. If you're running this against a cold list of 10,000 contacts and the prompt isn't perfectly calibrated, you're going to send a lot of mediocre long emails. Start with the shorter format and prove it works before scaling the longer version.

Video 2: Cold Email Is Hard Until You Build Systems Like This (Matt Lucero)

What the Video Is About

Matt Lucero has sent 10 million cold emails, generated over 10,000 interested leads, and booked more than 4,000 sales meetings across B2B industries. This video is a 37-minute breakdown of five principles for building cold email systems. The first section, on infrastructure, is where he spends the most time and where the real value is.

The core argument: most cold email failures aren't about copy. They're about infrastructure. Specifically, people take their main domain, plug it into a sending tool, and start blasting emails. The domain gets flagged, their outbound lands in spam, and then their actual transactional email starts landing in spam too. He tells a story of a lead whose invoices were going to spam because of this exact mistake.

The Infrastructure Framework

Matt's setup is straightforward but important. Instead of sending from one domain, you buy multiple domains, set up two inboxes per domain, and send a maximum of 15 to 25 emails per day per inbox. His recommended floor is 1,000 emails per day, which requires approximately 50 inboxes across 25 domains.

The math: 50 inboxes times 20 emails per day equals 1,000 daily sends. That's a reasonable minimum for getting statistically meaningful data on what's working.

He also talks about warmup. Every inbox you create needs to warm up before you send campaigns from it. Tools like Smartlead and Instantly have built-in warmup networks that trade emails between accounts on the platform, building sender reputation before you go live. His analogy is a new car showing up on traffic cameras a few times before starting to drive fast. Not perfect as analogies go, but the concept is right.

His sending tool recommendations: Smartlead as his top pick, followed by Instantly and a couple of others. He prefers Google and Microsoft inboxes for longevity and deliverability, and recommends buying domains through registrars and inboxes through resellers to reduce cost.

My Reaction

Everything he says about infrastructure is correct, and it's the part of cold email that still trips up smart people. I've talked to founders who have solid offers, clean copy, and good targeting, but they're getting 5% open rates because their main domain is cooked. They didn't protect it. They treated their sending domain like their brand domain and paid for it.

The speed limit analogy he uses is a good way to think about it. Google and Microsoft will technically allow you to send thousands of emails a day from a single inbox. They won't stop you. They'll just quietly route your messages to spam and let you wonder what went wrong. The system is designed to make lazy senders fail slowly.

My recommendation has always been to keep your main domain completely separate from outbound. Your outbound domains should be variations of your main brand, close enough to be recognizable but never the same. You can set those up easily through any major domain registrar and connect them to your sending tool in an afternoon. The full setup process is covered in my Cold Email Tech Stack guide if you want the step-by-step.

On the 1,000 emails per day floor, I mostly agree. At 500 emails a day, your results are going to be inconsistent. Some weeks you'll book five meetings, some weeks zero, and you won't be able to tell whether it's a copy problem, a targeting problem, or just variance. At 1,000 per day you start getting signal. At 2,000 to 3,000 per day you can run real split tests and iterate with confidence.

One thing he doesn't emphasize enough: email validation before sending. Sending to bad addresses destroys your bounce rate, and high bounce rates are one of the fastest ways to tank a domain. Before any list goes into your sending tool, it needs to be validated. ScraperCity's email validator handles this at scale without the per-email costs that add up fast on large lists. Clean list hygiene is the unglamorous part of infrastructure that nobody talks about in videos like this, but it matters as much as warmup.

I'd also add that the two-inboxes-per-domain recommendation is conservative but safe. Some operators run more than two and manage it fine. But if you're just getting started and you're not monitoring deliverability closely, two per domain gives you a buffer. You can always add more once you understand what your numbers look like.

What to Implement vs. What to Skip

Implement immediately if you haven't already: separate outbound domains from your main domain. This is not optional. One bad campaign on your main domain and you've got a company-wide deliverability problem. Implement: the horizontal scaling model. More domains, more inboxes, lower sends per inbox. It feels like more setup work upfront but it protects you and gives you far more sending capacity. Implement: warmup from day one. Any domain you create should be warming for at least two to three weeks before you send a single cold email from it.

The one thing I'd push back on is his framing around sending volume as the primary lever. He says "it's a numbers game" and implies you can compensate for other weaknesses by reaching more people. That's partially true. But I've seen plenty of campaigns that sent 5,000 emails a day and booked nothing, because the offer was weak or the targeting was off. Volume is a multiplier. It multiplies good strategy, but it also multiplies bad strategy. Fix the fundamentals first, then scale. For the fundamentals on the copy side, my Top 5 Cold Email Scripts is the right starting point.

The Pattern Across Both Videos

Here's what connects these two videos and why I picked them together this week.

Both Eric and Matt are making the same underlying argument from different angles. Cold email at scale is a systems problem, not a copywriting problem. Eric is solving it at the message level: use AI to generate genuinely relevant ideas so you don't have to manually personalize at scale. Matt is solving it at the infrastructure level: build horizontal sending capacity so you're not limited by inbox reputation or sending volume.

Most people focus on the email itself. They rewrite the subject line. They test the CTA. They obsess over the first sentence. And all of that matters. But if your infrastructure is broken, your best email doesn't get delivered. And if your angle is generic, your infrastructure delivers something nobody cares about.

The operators who consistently book 50 to 100 meetings a month have both dialed in. Clean infrastructure running at volume, pointed at a message that gives the prospect a genuine reason to respond. The creative ideas campaign is trying to solve the relevance problem at scale. The horizontal infrastructure model is trying to solve the deliverability problem at scale. You need both.

In The Cold Email Manifesto, I wrote about booking nearly 20 meetings from 60 emails in three days. The reason that worked wasn't magic copy. It was a tight offer sent to exactly the right people through a clean sending setup. The fundamentals Eric and Matt are covering have been true for a long time. They're just now being addressed with better tooling.

If you're building your list before you set any of this up, ScraperCity's B2B email database gives you verified contacts by industry and role without the noise you get from bloated databases. Clean input data is the third leg of the stool that neither video really covers. Good infrastructure, strong angle, clean list. All three.

The One Thing to Do This Week

If you're not already running the creative ideas format, build a version of it this week. Keep it simple. Write the email manually for five companies in your target market first. What are three genuinely useful campaigns you could run for each of them? Write those out. Then turn those five examples into a prompt template with the company description as the only variable. Test the one-idea-plus-P.S. format before the three-bullet version.

If your infrastructure isn't set up correctly, fix that first. No campaign performs if your emails aren't landing in the inbox. Get your sending domains separated from your main domain, get warmup running, and keep your daily sends at 15 to 25 per inbox. Then point that clean infrastructure at your creative ideas campaign and see what happens.

For follow-up sequences once you start getting replies, my Cold Email Follow-Up Templates give you the exact cadence I use across different verticals. The first response is rarely the close. The system around it is what converts.