07 / 07 · Lifecycle and email
Campaigns drafted, QA'd and held at the gate.
The email agent runs your segmented campaigns as a monthly loop. It refreshes and cleans the list, reviews last month's sends against the ledger, researches fresh material for each segment, drafts and QA's a campaign per segment, dry-runs the cohort sizes, and then stops. Nothing sends until a person approves it. It is the one channel where a mistake reaches everyone at once, so the gate is a feature, not a limitation.
What it reads
What it analyses
- How last month's campaign did per segment, against the baseline for that segment rather than a blended average.
- What happened after the click: pages visited, enquiries made, deals moved, read through the joined data layer.
- Which segments are tiring, from unsubscribes and falling engagement, before they go quiet.
- Whether the list is clean enough to send to, and cleaning it if not.
How it decides
One variable per segment per month, logged before drafting: the subject line approach, the opening, the offer, the send window. Verdicts are read at the enquiry end, not the open rate, using the campaign tag in analytics and the CRM. A segment that keeps losing gets its baseline and its assumptions rewritten rather than a fourth attempt at the same idea. Every campaign is dry-run for cohort size before it reaches the gate.
Example experiments
The shape of a test, written before it ships.
Illustrative, not a promise. Every experiment on your account gets this record before anything changes: the hypothesis, what the verdict is read from, what stays fixed, how long until a verdict, and where a person decides.
Hypothesis 01
A plain, personally-written opening outperforms a designed newsletter template on enquiries per send for the segment closest to buying.
- Measured on
- Enquiries attributed to the campaign tag, per recipient, against the segment's baseline.
- Held constant
- Send day, list and offer.
- Verdict after
- Two monthly sends.
- Human gate
- Both versions approved before the send.
Hypothesis 02
Sending to a segment defined by a real event in their world, such as a renewal or expiry, earns more enquiries than a calendar send to the whole list.
- Measured on
- Enquiries per recipient for the event-triggered segment against the calendar send.
- Held constant
- Copy approach and offer.
- Verdict after
- Two to three cycles of the event.
- Human gate
- You confirm the event and the data it is read from.
Hypothesis 03
Removing recipients who have not opened in six months lifts deliverability and engagement without lowering enquiries.
- Measured on
- Deliverability and enquiries per send before and after the clean.
- Held constant
- Content and cadence.
- Verdict after
- Two sends.
- Human gate
- Suppression list reviewed before it is applied.
Where people decide
Approval before every send. The agent never sends on its own.
You set the segments and the suppression rules.
Every draft is QA'd for links, tags, names and cohort size before it reaches you.
What you get
- 01A clean list and a segment review each month.
- 02One drafted, QA'd campaign per segment, waiting at the gate.
- 03Last month's results read at the enquiry end, per segment.
- 04A ledger of what each segment has responded to, so the next send starts from evidence.