All blog posts
article image

Outbound

DNS to SMTP: the outbound setup most agencies skip

article_pictureSamuel OdifaWriter for regulated industries, builder of agency growth systems

Most agencies that "do outbound" are running it on a domain someone set up in an afternoon, on records nobody has read since. The writing gets blamed when the numbers come back flat. The writing is usually fine. The setup underneath it was never finished.

This is the part that sits outside what a copywriter touches, which is exactly why it never gets done. Here is the whole chain, in the order it has to happen.

What has to be true before you send anything

Sending is the last step, not the first. Four things come before it, and skipping any one of them shows up as a deliverability problem two weeks later, when it is expensive to diagnose.

  1. A sending domain that is not your primary domain. If outbound burns a reputation, it should not be the reputation your invoices are sent from.
  2. Authentication records published and verified: SPF, DKIM, DMARC.
  3. Mailboxes created on that domain, one per sender, with real names on them.
  4. A warm-up schedule running for at least 3 weeks before the first campaign.

The records, in plain terms

Three records do the work. Every provider documents them differently and all three explanations assume you already know what the other two do, so here they are together.

SPF

A list of the servers allowed to send as your domain. It is a single TXT record, and the common failure is having two of them: a domain with two SPF records fails SPF entirely rather than merging them. If you are adding a sending tool, you extend the existing record. You do not publish a second one.

DKIM

A signature on every message, checked against a public key you publish in DNS. This is the record most often left half-finished, because the provider generates it and the person who generated it is not the person with DNS access. It sits in a shared doc for a month.

DMARC

The policy that tells receiving servers what to do when SPF or DKIM fails. Start at p=none and read the reports for 2 to 4 weeks before you move to quarantine. Publishing p=reject on day one is how agencies silently kill their own client onboarding email.

Warm-up is a schedule, not a feeling

Warm-up is where most of the improvised setups fall over, because it is the only step that cannot be compressed. A new domain has no history. You are building one, and the only input is consistent, low, human-looking volume over time.

  • Start at roughly 5 sends per mailbox per day and add about 5 a day.
  • Keep replies flowing. A conversation is the strongest positive signal you can generate.
  • Do not jump a cold mailbox straight to 50 a day because a campaign date moved.
  • Cap steady-state around 30 to 40 per mailbox. Volume comes from more mailboxes, not busier ones.
The agencies that get outbound working are not writing better emails than everyone else. They finished the setup, and then they wrote a decent email.

Where this actually goes wrong

Not in the technical steps. Each one is an afternoon of work. It goes wrong because the job spans three people who do not overlap: the strategist who sold the campaign, the copywriter who wrote it, and whoever holds the DNS login. Nobody owns the chain end to end, so it stalls at whichever step needs the other two.

That is a workflow problem wearing a deliverability costume. It is the same shape as the reporting problem I wrote about in the reporting workflow that stopped eating Mondays, and it gets fixed the same way: one person owns the sequence, and the sequence is written down.

If you want the version of this that runs without you, that is what lead-gen systems covers.

Bring me your messiest bottleneck

You talk, I listen and recommend a solution. Yes, I’d also tell you straight if I’m not the right fit.

Prefer async? Ask me anything →