Every rep sends from
their own mailbox.
Each BDR and AE connects their own inbox. The sequence sends through the pool of those mailboxes under a per-mailbox cap, and every positive reply lands on a named AE on the same event the classifier emits.
A shared sales alias has nobody behind it.
When a whole team sends from sales@ or sdr@, every BDR ends up sharing the same complaint history and the prospect has no name to reply to. A mailbox per rep means each one builds its own sending history and every reply comes back to a real person.
- Sent from by the whole SDR team
- Carries everyone's complaint history
- No named person for the prospect to reply to
- Belongs to one rep on the team
- Builds its own sending history
- Routes the reply back to a named AE
Comparing the two setups.
With a shared alias, every complaint and every reply belongs to no one in particular. With a mailbox per rep, both stay with the person they came from. It changes who handles a reply, who carries the reputation, and what happens when a rep leaves the team.
Sending limits the platform ships with.
Team volume scales with the size of the pool, not with the cap on any one mailbox. The defaults sit well under the 100 a day per-mailbox ceiling, so the team can compound throughput by adding healthy mailboxes instead of pushing any individual rep past their safe band.
- A mailbox at cap is skipped for the day, the campaign keeps moving against the rest of the pool.
- Per-mailbox reputation is isolated, so a watch-band flag on one rep does not slow the others down.
Most teams keep individual rep mailboxes around 50 a day on cold and grow team volume by adding mailboxes to the pool, not by raising the cap.
100 maxA 600 second gap between sends from one mailbox avoids the burst pattern provider heuristics associate with bulk sending tools.
every mailboxA mailbox drifting into the watch band has its volume halved and its spacing widened. A quarantined mailbox leaves the pool until it requalifies.
auto throttleClassification, the routing rule, the CRM stage move and the Slack notification all fire on the same inbound event, not as a chain of queued jobs.
one eventThe 50 / day figure is the platform default, allowed up to 100 in the configuration. Raising the cap is safe once a mailbox has gone a few weeks without complaints, bounces or watch-band flags.
Four routing rules, four fallback chains.
Pick one rule per sequence. The platform never silently changes who owns a reply. Every rule reaches an explicit terminal fallback, so a reply that does not match never disappears into a manager queue without a record of it landing there.
Next AE in the rotation. Quota skips on the same pass.
Match on recipient domain or contact territory field.
CRM owner attached to the contact gets the assignment.
Manager queue with an SLA timer. Bulk re-assign.
Plain text emails.
When a BDR writes to one prospect, they do not add a banner or a tracking pixel. Those are the signals filters use to classify marketing mail, so a cold email carrying them tends to land in Promotions. Plain text reads like a normal note and is more likely to reach the main inbox.
The deck lives on a hosted page, so the email stays light and reads like a note from a person.
Personalisation with Go templates.
Every step is a Go text/template, the same syntax used across the rest of Warmbly. Pull in a prospect's details with a leading dot, branch on a field with if / else, and the template renders per prospect at send time.
- An empty field renders as nothing, so a missing trigger or company never leaves a broken sentence behind.
- Wrap an optional detail in an if, and it only appears when the field is set on the contact record.
{{.FirstName}}{{.Company}}{{.Role}}{{.Industry}}{{.RecentEvent}} Her RecentEvent was blank, so the if skips the "at" clause and the sentence flows without a stray gap.
The CRM stays the source of truth.
Two-way sync with HubSpot, Salesforce or Pipedrive on contacts, deals, stage and owner. The activity feed is outbound only, so the CRM is never fought over. A stage moved over there is honoured here on the next pass.
- An assignment writes the owner, the stage and the activity on the same event the Slack ping fires from.
- Webhooks are HMAC-signed and idempotent on (message-id, rule version).
ContactDealStageOwnerActivityNote The CRM picks up the stage change, the owner and the activity off the same event the Slack ping fires from.
One reply, every surface updated at once.
When a positive reply arrives, the assignment, the Slack ping, the CRM stage move, the deal note, the Unibox claim and the webhook all fire from the same event. The AE picks up the conversation on a single notification, and the rest of the surfaces are already in sync by the time they look.
"Yes, this is timely. Could we put 20 minutes on the calendar next week?"
Lands on the AE with the booking link in the auto-acknowledgement and the sequence paused for the contact.
Picked up by the AE for nurture. The sequence holds for 60 to 90 days, then resumes from the next step.
Workspace suppression on the contact, the opt-out lands on the record, and the deal closes as lost-no-fit.
Each reply class has its own path. The fan-out above is the positive case, the strip names the three other classes you will see most often.
A few questions sales ops asks first.
Add the rep mailboxes to a sender pool. The scheduler round-robins sequence sends across the pool with awareness of each mailbox cold cap, the 600s spacing rule, and the current health band. A mailbox that hits its cap is skipped for the rest of the day, the campaign keeps moving against the healthy mailboxes that still have headroom.
Replies are classified on arrival and routed by the rule on the sequence. The assignment lands as a Slack notification, a CRM stage move, and an Unibox claim for the named AE, all on the same event. The originating sequence is paused for that contact at the same moment.
Per-mailbox reputation is isolated. A mailbox that drifts into the watch band has its sends throttled and spacing widened. If it crosses into quarantine it leaves the sender pool entirely until it requalifies. The rest of the pool keeps sending at full pace.
Set up a mailbox for every rep on the team.
Connect each BDR and AE mailbox, drop them into a sender pool, and route every positive reply on the inbound event.