Cold email works when it is a short, honest note to one person who has a problem you can solve. It fails when it is a broadcast. This chapter explains why the channel is still worth your time, what changed in the last two years, and the five layers every working program is built on.
The short answer
Every year someone declares cold email dead. What actually died is the lazy version: one template, a bought list, ten thousand sends a day from a brand new domain. Mailbox providers got better at spotting that pattern, and buyers got better at ignoring it. The careful version, a relevant message sent to a small list of the right people from a sender that is set up properly, still opens doors that ads and social posts cannot.
Email has three things no other outbound channel has at the same time. It reaches a named person, not a feed. It is asynchronous, so the reader answers when it suits them. And it is cheap to send and cheap to test, so you can learn which message works for which buyer within a couple of weeks. Phone calls interrupt. Social messages compete with a hundred notifications. Paid ads rent attention you have to keep paying for. Email, done well, is the only one where a stranger can reply in their own words, on their own time, with a real question.
Our position
We think cold email should be the first outbound channel for most B2B teams selling something with a clear buyer. Not the only one. The first, because it teaches you fastest what your market cares about. Chapter 10 covers how to add LinkedIn and phone on top.
What changed, and why it helps careful senders
In February 2024 Google and Yahoo started enforcing sender rules that used to be soft advice. In May 2025 Microsoft followed for Outlook.com, Hotmail and Live addresses. The details are in Chapter 9, but the effect is simple: the bar for getting into the inbox went up, and it went up most for people who send a lot of mail they cannot justify.
Read those rules as good news. They push out the senders who were filling inboxes with noise. If you authenticate your domains, send modest volumes, and write to people who might actually want the message, you are now competing with fewer voices. The rules punish volume without relevance. They do not punish a well targeted note.
Two other things changed. First, open tracking stopped being a reliable signal. Apple Mail Privacy Protection loads images on the reader's behalf, so a tracked open often means nothing. Google says plainly that it does not track open rates and cannot verify open rates reported by third parties. We cover what to measure instead in Chapter 12. Second, AI made it trivial to produce a thousand slightly different emails. That raised the floor of personalisation, so a merged first name and company name no longer reads as effort. What reads as effort now is a reason: why this person, why now.
The five layers of a working program
Every cold email program that keeps working for more than a quarter rests on the same five layers. Each layer depends on the one under it. When results drop, the cause is almost always in a lower layer than the one people start fixing. Teams rewrite subject lines when the real problem is that their domain has no DMARC record, or that half the list left their jobs last year.
Infrastructure. Separate sending domains, authenticated mailboxes, warm-up, and sane daily caps. If this layer is wrong, nothing above it gets seen. Chapters 3, 4 and 9.
Use the stack as a checklist before you write a single word. If you cannot point to a sending setup, a verified list, and a reason for each account, you are not ready to write copy yet. Copy is the most visible layer, which is why people over-invest in it, but it carries the least weight.
Who cold email is for, and who should skip it
Cold email fits best when four things are true. You can name the job title that feels the pain. You can find those people and their companies with public information. The problem you solve is costly enough that a stranger will spend two minutes on it. And a first conversation is a reasonable next step, rather than a purchase.
| Situation | Good fit? | Why |
|---|---|---|
| B2B service or software with a clear buyer role | Yes | You can find the buyer and speak to a specific pain. |
| Agency or consultancy selling to a defined niche | Yes | A narrow niche makes research fast and messages sharp. |
| Recruiting, partnerships, investor or podcast outreach | Yes | One-to-one asks with an obvious reason to reply. |
| Low-priced consumer product | No | The deal is too small to justify personal outreach, and consumer inboxes have stricter expectations. |
| A product nobody has bought yet, with no clear buyer | Not yet | Do customer interviews first. Cold email is a good way to book them, not to sell. |
| Regulated offers that need prior consent in your market | Check first | Some countries require consent before any commercial email. See Chapter 13. |
Where replies come from
A cold email program is a funnel, but not the one most dashboards draw. The useful version starts with the market you could reach, narrows to the accounts that fit, then to the people at those accounts with a live reason to talk, then to the ones who reply, and finally to the conversations that turn into meetings. Every stage leaks. Your job is to make sure the leak happens early and cheaply, at the research stage, instead of late and expensively, after you have burned a domain on people who were never going to buy.
Reachable market. Everyone you could legally and practically email. Big, and mostly the wrong people. Do not send to this.
Notice that the biggest decision happens before any email goes out. Teams that filter for fit and for a live reason send far fewer emails and get more conversations from them. Teams that skip that step send more, get flagged more, and conclude that cold email does not work.
Myths worth dropping
- More volume means more meetings. Past a modest daily volume per mailbox, more sends mostly mean more spam complaints. Scale with more mailboxes and better lists, not with bigger daily numbers per box. Chapter 14.
- Personalisation means a custom first line. A compliment about someone's LinkedIn photo is not personalisation. A reason that only applies to them is. Chapter 7.
- A clever subject line is the key. Subject lines that sound like marketing get filtered by people faster than by algorithms. Plain and specific wins. Chapter 6.
- Open rates tell you if it is working. They no longer do. Track replies, positive replies and meetings. Chapter 12.
- AI will write it all for you. AI is very good at research and first drafts. It is bad at knowing which facts matter to a buyer. Keep a human reading every message before it goes out until you trust the pattern. Chapter 11.
- Cold email is illegal. In most markets, B2B cold email is legal with conditions: honest headers, a clear sender, a working opt-out, and in some countries a legitimate interest or prior consent. Chapter 13 covers the rules that matter.
What a good first month looks like
The first month is not about revenue. It is about building a setup that will not break, and learning which message lands with which buyer. Here is the order we recommend. Each step is covered in depth later in the course.
Week 1: set up sending
Buy two or three domains close to your brand, create two or three mailboxes on each, add SPF, DKIM and DMARC, and start warm-up. Do not send cold mail from your main domain. Chapter 3.
Week 1 and 2: define one buyer
Write down one ideal customer profile: industry, size, the role that owns the problem, and the trigger that makes the problem urgent. One. Not five.
Week 2: build a small list
Find around two hundred fit accounts and one or two people at each. Verify every address. Note one live reason per account. Chapter 5.
Week 3: write two angles
Write two different first emails for the same buyer, each built around a different pain or trigger. Keep both short. Plan three or four follow-ups each. Chapter 6.
Week 3 and 4: send slowly and read every reply
Start with a small daily number per mailbox and raise it gradually. Read every reply yourself, including the angry ones. Sort them into interested, not now, wrong person, and no.
End of month: keep the winner, change one thing
Compare the two angles on positive replies, not opens. Keep the stronger one and change a single variable for the next batch: the trigger, the offer or the first line.
The one mistake that costs the most
Sending cold email from the domain your company uses for customer, billing and team mail. If that domain's reputation drops, your invoices and password resets start landing in spam too. Always use separate sending domains.
How this course is organised
The fifteen chapters follow the stack from the bottom up. Chapters 2 to 4 cover mindset and the sending setup. Chapters 5 to 8 cover lists, research, writing and follow-up. Chapters 9 to 12 cover deliverability, other channels, AI and measurement. Chapters 13 to 15 cover the law, scaling up and the advanced plays most teams never try. Every chapter has a figure you can click through, and every number that is not advice comes with a link to the original source so you can check it yourself.
If you are short on time, read Chapters 3, 6 and 9 first. They prevent the most expensive mistakes. If you already run cold email and results have dropped, start with Chapter 9, then come back to Chapter 5. Most drops come from deliverability or list quality, not copy.
Where FirstSales fits
FirstSales is the tool we build for this exact workflow: connect sending mailboxes, build a list, research each account, draft emails a person can review, and run follow-ups that stop when someone replies. You can do everything in this course with other tools or by hand, and we say so where it matters. Where FirstSales saves time, we point to the demo so you can see it with sample data before you decide anything.
Key takeaways
- Cold email is still the fastest way to start a conversation with a named B2B buyer.
- New sender rules from Google and Microsoft punish volume without relevance. Careful senders benefit.
- Build from the bottom: infrastructure, list, reason, message, follow-up.
- Filter hard before you send. Small, well researched lists win.
- Measure replies and meetings. Open rates are no longer reliable.
- Never send cold mail from your main company domain.
Respostas rápidas sobre a masterclass
Quantos e-mails frios devo enviar por dia?
Comece com 5-10 por caixa de entrada por dia em contas novas. Após 2-3 semanas de aquecimento, suba para 35-50 por caixa. Se precisar enviar 500+ e-mails por dia, use 10-15 caixas aquecidas e alterne os envios entre todas elas. Nunca force uma única caixa além de 50 e-mails frios por dia.
Qual é uma boa taxa de resposta no e-mail frio?
Não existe um número único e honesto. A taxa de resposta depende do seu mercado, da sua lista, da sua oferta e de como as respostas são contadas, então números de campanhas alheias raramente se aplicam. Compare cada campanha com os seus próprios resultados anteriores para o mesmo tipo de público e avalie pelas respostas interessadas, não pelo total. O capítulo 12 mostra como.
Qual deve ser o tamanho de um e-mail frio?
Mantenha os e-mails frios entre 50-125 palavras. Alguns estudos sugerem até 150 palavras para certos públicos B2B. E-mails mais curtos têm desempenho consistentemente melhor. Se você precisa de mais de 125 palavras para expor sua ideia, sua mensagem não está focada o suficiente.
Devo usar HTML ou texto puro nos e-mails frios?
Texto puro. Sempre. E-mails em HTML com logos, imagens e layouts formatados acionam filtros de spam e parecem e-mails de marketing. Um e-mail em texto puro de uma pessoa para outra é exatamente como um e-mail frio deve parecer. A única exceção é a sua assinatura, que pode incluir formatação básica.
Quantos follow-ups devo enviar?
Normalmente alguns poucos, com vários dias de intervalo, cada um trazendo algo novo: outro ângulo, um recurso útil ou uma pergunta simples. Pare assim que alguém responder e termine com um encerramento curto e educado. O capítulo 8 trata do ritmo e do que dizer em cada etapa.