Why separate the sending domain from my brand domain?

Because every email that lands in spam damages that domain's reputation, and if it is the same domain you use for proposals and invoices, the damage reaches your working email. The standard practice is a separate sending domain that redirects to the main one.

Is this audit useful if I do not do cold outreach?

Partly. The infrastructure checks — SPF, DKIM, DMARC, sending domain, Postmaster Tools — apply to any automated email, including transactional and follow-up messages. The list and frequency checks are specific to outbound prospecting.

07 · Outbound

Why do your cold emails land in spam?

Twenty checks across infrastructure, list, sending and content. The diagnosis comes out ordered by impact: the first item is what costs you most, not what is easiest to fix. If the first line says you send from your main domain, none of the other nineteen matter until that is resolved.

No signup · no email · runs in your browser

Infrastructure

7 checks

Which domain do you send cold email from?

How does the mail leave?

Do you have SPF published?

Is DKIM signing your mail?

And DMARC?

Did you warm up the domain before sending?

Is your domain registered in Google Postmaster Tools?

The list

5 checks

Do you verify addresses before sending?

What share are role addresses (info@, contact@)?

What is your bounce rate?

Do you auto-suppress hard bounces?

Do you dedupe by address across campaigns?

Sending

4 checks

How many emails does each mailbox send per day?

Is there a lifetime cap per address?

Do you include List-Unsubscribe?

Which folders does your bounce and reply scanner read?

Content

4 checks

Do "free" or "no cost" appear in your templates?

How long are your emails?

Do you put links in the first email?

How personalised is each email?

Why it exists

This audit comes from my own mistakes

I ran the full diagnosis on 814 of my own emails. The result was a 0.25% reply rate, far below market average. The intuitive conclusion was "the copy is wrong". It was false.

The placement test sent three real emails to Gmail and all three landed in spam, including a control without List-Unsubscribe. SPF, DKIM and DMARC all passed. None of the 24 sending IPs appeared on six blocklists. Mail-tester gave 10/10 to the same email Gmail sent to spam. With 100% spam placement, that reply rate was not measuring template quality: it was measuring how many people check their spam folder.

The hardest finding: the bounce scanner read only INBOX and structurally reported 0.00%. The real mailbox had 1 message in INBOX and 41 in the trash — everything was read and deleted from the phone before the script ran. Once fixed, 14 bounces appeared: 3.43%, above the threshold where the standard recommendation is to stop sending.

That question is check 16 in this audit, and it is there because it took me months to find.

Questions

What people usually ask

Because perfect authentication is not deliverability. All three can pass, the Return-Path can align, there can be no HTML and no links, and a neutral analyser can give the message 10/10 — and Gmail will still send it to spam. When that happens there is nothing left to fix inside the message: what is being penalised is domain reputation, and it does not recover while you keep sending from it.

Other tools

General assessment

Did you get a number you did not expect?

Send it over and I will tell you what I would do with that case in the first 30 days, with nothing to hire. If the honest answer is that you do not need an agency yet, I will tell you that too.

Let’s talk See the service