spaceout.

2026-09-10

The 30 day cold email setup most agencies skip

Cold email fails for two boring reasons. It never reached the inbox, or it read like a tool wrote it. The second problem gets all the attention because copy is visible and infrastructure is not. But the first problem is decided in the first month, before anyone judges a subject line, and it is the part most agencies skip because it does not look like progress.

This is what that month actually contains, week by week. Use it as a playbook if you are building outbound in house. Use it as an interview script if you are evaluating an outbound lead generation agency for your SaaS company. Either way, if the plan in front of you promises meetings in week one, the plan in front of you is skipping this.

Why the first month has almost no sends in it

The outcome you are buying in month one is not meetings. It is a sending stack that is safe to scale, with your primary domain nowhere near the blast radius.

Mailbox providers score senders on history. A brand new domain with no sending history that suddenly pushes cold volume looks exactly like what it is, and gets filtered accordingly. An agency that starts blasting in week one is spending your domain reputation to hit an early activity number, and the bill arrives quietly: placement degrades, replies dry up, and because opens and sends still get reported, nobody notices. Plenty of founders have a version of the story where their email went to spam and nobody caught it for weeks.

The fix is sequencing. Infrastructure first, warm-up next, volume last. Here is the sequence.

Week 0: buy the domains your primary will never risk

Before anything else, buy separate sending domains. Not subdomains of your main domain, and never the main domain itself. Lookalike domains work well: getyourcompany.com alongside yourcompany.com. You should own them, whoever sets them up. If an agency registers them under its own account, you have a hostage situation waiting for the day you part ways.

The reason is containment. Cold outreach carries deliverability risk no matter how well it is run. Separate domains mean that risk lives somewhere you can afford to lose, while the domain your team uses for customer email, billing and support stays clean.

Then configure authentication on each sending domain:

SPF tells receiving servers which services are allowed to send on the domain's behalf. DKIM signs each message so the receiver can verify it was not tampered with and genuinely came from your domain. DMARC tells receivers what to do with mail that fails those checks, and gives you reports on who is sending as you.

All three, set up correctly, on every sending domain, before a single message goes out. This is table stakes, and it is checkable: any agency you are evaluating should be able to show you the records on a live client setup. While the DNS propagates, provision the inboxes on those domains, real accounts with names, signatures and sensible sending identities.

Weeks 1 to 3: warm the inboxes before you trust them

A configured inbox is not a trusted inbox. Warm-up is the process of giving new accounts a believable history: gradually increasing send activity, messages that get opened and replied to, behaviour that looks like a person doing a job rather than a script doing a campaign.

During this period the inboxes send and receive at low, slowly rising volume. Nothing goes to prospects. The goal is that by the time real outreach starts, mailbox providers have seen these accounts behave like real accounts for long enough to extend them the benefit of the doubt.

Two honest caveats. First, on duration: the weeks in this walkthrough are a planning frame. In our own programmes, inboxes warm over [CONFIRM: number] weeks before meaningful volume, and rushing that number is precisely the corner-cut this post is about. Second, warm-up is not fire and forget. Inbox placement should be monitored throughout, not checked once at setup, because a warming account that starts landing in spam needs intervention before launch, not a post-mortem after it.

If a vendor cannot tell you how long they warm, how they monitor placement during warm-up, and what they do when placement slips, you have learned what you needed to learn.

Weeks 1 to 3, in parallel: build the list that makes the copy work

Warm-up time is not dead time. While the infrastructure matures, the research runs.

A precise list makes mediocre copy work. A bad list makes great copy irrelevant, and it does something worse: bounces and spam complaints from a scraped list damage the very sending reputation you just spent weeks building. List quality is a deliverability input, not just a targeting choice.

Building the list properly means starting from a written ICP definition, not a job title search. It means verifying every address before it enters a sequence, because unverified addresses bounce and bounces poison placement. It means deduplicating against your CRM and against suppression lists, so existing customers, open deals and people who have already opted out never receive a cold email. And it means enriching each account with the context that decides whether this month is the right month to reach out: funding, hiring, product launches, leadership changes, tech stack.

The volume trade-off is real and you should make it deliberately. Two hundred researched accounts will outperform five thousand scraped ones on almost any serious B2B motion, and on a narrow enterprise ICP the researched list is the only version that works at all.

This is also where compliance gets built in rather than bolted on. Legitimate-interest B2B data, identification and unsubscribe in every send, and opt-outs suppressed immediately and permanently. An agency running many clients should suppress opt-outs across all of them, not per engagement.

Week 4: first real sends, at deliberately low volume

Now the sequences go live, and the discipline continues. Start at low daily volume per inbox and ramp gradually, watching placement and reply data as you go. Volume is the last lever you pull, not the first.

The copy itself should be written per segment, not per company: short, specific, one idea per email, a clear reason for the outreach and an ask that is easy to say yes to. You should see and approve everything before it sends. And from the first send, replies need an owner. Interested prospects get qualified and booked. Not-now prospects get dated and nurtured. Wrong-person replies get routed to the right person in the account. Opt-outs get suppressed the moment they arrive.

After week 4: deliverability is a job, not a setup task

The setup month ends. The monitoring does not. Placement, bounce rates, reply rates and spam complaints need watching every week, because sending reputation is earned continuously and lost quickly. Weekly reporting should show sends, deliverability and placement, open and reply rate, positive reply rate, and meetings booked and held, split by segment, along with what changes next week and why. If the report only shows opens, the report is hiding something.

The evaluation checklist

Whether you are shortlisting a B2B cold email agency in India or comparing global vendors, the same questions expose the same corners: Whose domains do the emails send from, and who owns them? Show me SPF, DKIM and DMARC on a live setup. How long do you warm, and how do you monitor placement during and after? How is every address verified, and what gets suppressed? What happens to replies, including the negative ones? An agency that has done this work answers in specifics. An agency that skipped it changes the subject to meeting guarantees.

If you would rather see how we run this setup for funded AI and B2B SaaS companies entering new markets, book a call and bring the market you are trying to win.

Tell us which market you are trying to win.