what does it take to stand up a cold email programme from nothing and keep it sending every week?

Blooming Bay outbound programme — sequence and stack overview

Context

Blooming Bay needed outbound that ran on its own. Not a list bought once and blasted, and not a tool licence with nobody behind it, but a sending operation that could be handed over and left to work.

I set the whole thing up from scratch and have run it since, on the same stack I use for my other outbound clients. Domains, mailboxes, warm-up, lists, sequences, and the routing that gets a reply in front of a human quickly.

The setup

The first month of any outbound programme is not campaign work, it is plumbing, and skipping it is the most expensive shortcut available. I buy dedicated sending domains, provision the mailboxes in Mailforge with the authentication records set properly, and put every one of them through Warmforge before it sends anything real.

Nothing sends from the company's primary domain. Volume stays low per mailbox and spreads wide across the pool, which is slower to ramp and far harder to break.

  • Dedicated domains, bought and configured for outbound only.
  • Mailboxes provisioned with authentication correct from day one.
  • Full warm-up cycle before the first real send.
  • Conservative per-mailbox volume, spread across the pool.

Lists and sequences

Lists are built and enriched in Leadsforge, then filtered down before anything is written. Sequences run six touches in Salesforge, split across email and LinkedIn so the second message arrives to a name the reader already half recognises.

Every variant goes through the same checks before it ships. A banned-phrase sweep to remove the language that marks a message as bulk, and a rendered preview with real merge fields so no one ever receives a bracket where their company name should be.

Replies and routing

Replies are pushed to Slack by webhook the moment they land, with enough context attached that they can be answered without opening the campaign tool. Contacts are enriched and pushed onward to the CRM automatically, so the follow-up does not depend on anyone remembering to log it.

what actually matters

Reply speed beats reply volume. A message answered the same hour is worth several that sit unread until Friday.

Running it week to week

Setup is the easy half. The ongoing work is watching for the things that drift: a mailbox whose placement has slipped, an angle that stopped landing, a list segment that was never as good as it looked. Volume gets adjusted, weak rows get cut, and copy gets rewritten when the numbers say so rather than on a schedule.

Stack

The point of a programme like this is that it stops being a project. It sends whether or not anyone is thinking about it, the replies find a human quickly, and the only thing left to do each week is the judgement: what to push harder on, and what to stop.

Next Project
Magneto outbound programme