Cold Email Foundations Most People Skip

What I Watched This Week

I watch a lot of cold email content. Most of it is recycled advice from people who haven't sent a campaign in months. But two videos stood out this week because they both hit the same nerve from completely different angles: the stuff that happens before you write a single word of copy.

One is about who you target and how you think about outbound. The other is about the bare minimum infrastructure you need to get emails delivered. Together, they cover the two places I watch cold email campaigns fall apart.

Video 1: The Case for Manual Outbound First

This is a solid 12-minute breakdown from a visiting partner at YC who has helped hundreds of batch companies fix their outbound. The core framework is eight tips for improving reply rates, but the value is concentrated in the first three.

Start Manual Before You Automate

The first tip is one I've been preaching since I wrote The Cold Email Manifesto: do at least 100 outreaches by hand before you automate anything. The video makes the point well. When founders load up a tool, blast a thousand emails, and get zero replies, they've learned absolutely nothing. Was it the messaging, the targeting, the subject line? Could it be deliverability? They have no idea.

I agree with this completely. When I was building my first cold email systems, I personally wrote and sent thousands of individual emails. That manual process taught me which phrases got replies, which job titles responded, and which objections kept coming up. You cannot shortcut that education.

The video puts it this way: "Spend time thinking about one individual person. What is her problem specifically? What would capture her attention? Visit her LinkedIn page and read about her company." That homework is how you generate your highest-converting outbound ideas.

Founders skip straight to automation because it feels productive. Sending 500 emails in a day feels like work. But if those 500 emails are untested garbage aimed at the wrong people, you just burned 500 leads you can never email again with a fresh first impression.

Targeting Beats Perfect Messaging

This is the tip I'd highlight if I could only pick one from the entire video. The example given is a YC company selling an infrastructure product to senior engineering leaders. But when the founder's prospect list was audited, it was full of DevRel managers, GTM managers, even a customer success manager. Hundreds of messages sent to people who would never buy, no matter how good the email was.

People will spend hours tweaking a single sentence in their email body while their list is 40% wrong titles. A mediocre email to the perfect prospect will outperform a perfect email to the wrong person every single time.

The video introduces a useful framework for finding the right targets: look at who signed the contract and paid in your existing deals. Look at who has already replied to your outbound. And if you have neither of those yet, ask yourself who in the organization has the most to gain from buying your product.

The example used is a security monitoring tool. A chief security officer might get fired if there's a data breach, so they're highly motivated. An engineering manager would probably rather not be monitored at all. Same product, completely different levels of interest based on job title alone.

This is exactly why I tell people to start with a proven script framework and then spend the majority of their time on list quality. Your scripts matter, but they matter a lot less than who receives them.

Intent Signals and the Needle in the Haystack

The video also covers intent signals, and this is where it gets useful for intermediate cold emailers. The example: if you sell compensation planning software, companies below 100 employees probably aren't introducing salary bands yet. Companies above 500 already have them locked in. The sweet spot is companies crossing that 100-employee threshold.

Every product has signals like this. Job postings are called out as a great source. If a company is hiring for a role that relates to the problem you solve, that's a strong indicator they're feeling that pain right now.

This is something I've built entire campaigns around. When I was generating millions in revenue through founder-led sales, some of my best-performing campaigns were triggered by specific hiring patterns or growth milestones. The evidence is right there in the job posting or the headcount graph.

To find these signals at scale, you need good data sources. Tools like ScraperCity's B2B database let you filter by company size, growth rate, and other attributes that map directly to intent. The alternative is manually checking each company, which works at small scale but doesn't hold up past your first 200 prospects.

The Email Craft Section

The video's messaging advice is solid but not groundbreaking. Keep it short (five to eight sentences for email, two to five for LinkedIn). Lead with the prospect's problem, not your product features. Include a specific call to action like "Got 20 minutes to hop on a call this week?" instead of "Does this sound interesting?"

One thing I appreciated: the video shows a cold email from a YC founder selling an SMS conversion product. The email opens with "I used to be a software engineer at Tesla. We got very good at using SMS to convert the people who visited our website into car buyers." That's a strong credibility play. It's informal, peer-to-peer, and immediately establishes why the sender is worth listening to.

The call to action at the end asks if the prospect would be open to "an experiment" to see if they could improve conversion rates. Framing it as an experiment instead of a commitment lowers the bar significantly. Smart move.

The video also briefly touches on subject lines, noting that creative and slightly weird ones often outperform descriptive ones. The example: "broken agents" outperformed "better agents for every workflow." Same thing in my campaigns. If you want more ideas for what works in subject lines, I put together a resource for that at /subject.

What to Implement

Implement the manual-first framework immediately if you haven't done it. Even if you've been running automated campaigns for a while, going back to manual for a new ICP or a new offer will teach you things your analytics dashboard never will.

The targeting audit is also immediately actionable. Pull up your current prospect list right now. What percentage of those people have the pain point you solve? What percentage have the authority to buy? If you can't answer both questions with confidence for at least 80% of your list, stop sending and fix your list first.

The LinkedIn profile section (tip four in the video) and the reply rate debugging order (tip five) are useful but less critical. Optimize your LinkedIn after you've nailed targeting and messaging, not before.

Video 2: The $50/Month Minimum Viable Stack

This video takes a very different approach. Instead of strategy, it's focused on infrastructure: what's the absolute cheapest cold email setup you can run? The format is a daily updates style with community Q&A, so the cold email stack breakdown is embedded within a longer video. I'll pull out the relevant parts.

The Minimum Stack Breakdown

The recommended stack breaks down to roughly $50/month:

• Lead sourcing: Apify with Apollo-based databases, roughly $1.50 to $2 per thousand leads
• Email verification: NeverBounce, Million Verifier, or Mail.so to clean the list
• Mailboxes: Zap Mail at about $3/month per mailbox (a few mailboxes needed)
• Domain: About $1/month ($10/year)
• Sending tool: Instantly at $37-39/month

Total comes to roughly $50/month for sending about 1,000 emails per month (roughly 30/day).

The video is upfront that 1,000 emails per month is quite low. The exact quote: "Obviously you're going to want to send more than a thousand a month. You know, 30 a day. Otherwise, why the hell are you doing cold email in the first place?" Fair point. But as a proof-of-concept budget, it works.

Where I Agree

The emphasis on verification is correct. When you're sourcing leads from aggregated databases, bounce rates can be high. Sending to bad emails tanks your sender reputation fast, and once your domain is burned, you're starting over. A few dollars spent on verification saves you hundreds in burned infrastructure.

I also appreciate the honesty about Zap Mail having "had some issues in the past." Too many creators pretend every tool they recommend is perfect. Mailbox providers for cold email change all the time. What works today might get cracked down on tomorrow. That's reality.

The core insight is right: cold email infrastructure is cheap. Knowledge is the barrier to entry. Knowing how to set up authentication, warm up domains properly, write emails that don't trigger spam filters, and build lists that don't bounce. If you want to see what a more complete tech stack looks like once you're past the minimum viable stage, I break that down at /coldemailtechstack2025.

Where I'd Push Back

The lead sourcing recommendation gives me pause. Using Apify to access Apollo-type databases at $1.50-2 per thousand sounds great on paper. But leads at that price point come with quality problems - outdated titles, wrong companies, dead addresses. The video acknowledges this ("pretty high bounce rate, so be quite careful"), but I'd go further: cheap leads often have outdated job titles, wrong companies, and email addresses that technically verify but belong to people who left six months ago.

This connects directly to what the first video was saying about targeting. If you save $20 on leads but half your list is wrong titles or dead emails, you haven't saved anything. You've wasted your budget on bad contacts.

For lead sourcing, I'd recommend investing a bit more in data quality upfront. Tools like ScraperCity's email finder or Findymail give you higher-quality contacts that match your ICP. You can also use ScraperCity's email validator to clean whatever list you build before it hits your sending tool.

I'd also note that at 30 emails per day, you're operating at a volume where manual outreach (as recommended in the first video) might be more effective. At that scale, you could personalize every single email with specific research about the prospect's company. The ROI per email would be dramatically higher than a templated blast to a cheap list.

The Maker School Advice

There's a section in the video where a 17-year-old from Pakistan asks about joining a coaching program with only one month of savings. The advice given is to save up for at least three to four months before joining, because not everyone gets a client in their first month, and there are additional costs (mailboxes, tools, etc.) that eat into your budget.

I respect this. It would have been easy to take the enrollment. Instead, the advice was practical: you're young, you can't use platforms like Upwork or Fiverr at 17, and one month of runway isn't enough to give yourself a fair shot. That's honest guidance, and it's the kind of thing I wish more people in this space would say.

What to Implement

The $50 framework is useful as a mental model for what the absolute floor looks like. If someone tells you cold email requires $500/month in tools before you can send your first campaign, they're either overcomplicating it or trying to sell you something.

The verification step is non-negotiable regardless of budget. Build that into your process from day one.

But don't let the low price fool you into thinking cheap infrastructure equals good results. The stack is the easy part. Targeting is hard. Writing messages people want to read is hard. And you still have to iterate based on reply data. The first video covers all of this.

Two Failure Modes

These two videos represent the two failure modes in cold email.

Failure mode one: Great infrastructure, terrible targeting. You've got your domains warmed up, your sending tool configured, your sequences loaded. But your list is garbage and your message is generic. This is the founder from the first video blasting a thousand emails and learning nothing.

Failure mode two: Great targeting instincts, terrible infrastructure. You know exactly who to email and what to say, but your emails land in spam because you skipped authentication, used a burned domain, or sent from a provider that's already blacklisted. This is the commenter from my newsletter who was using Mailgun for cold email and getting 50% open rates with zero replies.

Neither targeting nor infrastructure works alone. You need both. And critically, you need them in the right order.

My recommended sequence:

1. Define your ICP with specificity (job title, company size, industry, intent signals)
2. Build a small, high-quality list of 100-200 prospects manually
3. Set up basic infrastructure (verified domain, authenticated mailbox, warm-up period)
4. Send 100 manual, personalized emails
5. Analyze what worked and what didn't
6. Only then automate what's proven

This sequence combines the best advice from both videos. Manual targeting discipline from the first. Infrastructure awareness from the second. And a feedback loop that teaches you something before you scale.

What to Do This Week

If you're getting zero or near-zero replies right now, do not touch your email copy. Instead, do these three things:

First, audit your list. Open your current prospect list and check 20 random entries. For each one, ask: does this person have the pain point I solve? Do they have authority to buy? Is their email valid? If more than 3 out of 20 fail these checks, your list is the problem.

Second, check your infrastructure. Send a test email to a Gmail account you own. Does it land in Primary, Promotions, or Spam? Check your SPF, DKIM, and DMARC records. If any of these are misconfigured, fix them before doing anything else.

Third, send 10 truly manual emails. Pick 10 prospects from your cleaned list. Visit each one's LinkedIn profile and company website, then write each email from scratch with a specific reference to their situation. Track which ones reply. Those replies will teach you more about your market than 1,000 automated sends ever could.

The founders who generate pipeline from cold email did the boring work of understanding their buyer before they ever hit send.