Small marketing teams running Mailchimp campaigns rarely have a copy editor on staff. An agent can draft the subject line, preheader, and body from a brief in seconds, but Mailchimp itself does not record who drafted, who reviewed, or why a send went out. Dock fills that gap. The agent writes the campaign to a Dock copy brief, a named editor approves the row, and only then does the campaign push to Mailchimp. The artifact lives in Dock with attribution attached.
Mailchimp and HubSpot stay the system of record for the raw data. Dock is the system of record for what the agent interprets. Each Dock row carries a pointer back to the platform record, agent identity, decision, reviewer, and timestamp. The agent re-fetches platform data via fresh API reads when it needs current state.
The Campaign Briefs table
| Campaign ID | Mailchimp link | Segment | Agent draft (subject) | Drafted by | Editor | Status | Sent at |
|---|---|---|---|---|---|---|---|
| CMP-2041 | mc/campaigns/2041 | Active trials, day 7 | "Three things to try before Friday" | mei-marketing-agent | govind@ | Approved | 2026-05-28 09:14 |
| CMP-2042 | mc/campaigns/2042 | Churned, 30-day winback | "We rebuilt the part you flagged" | mei-marketing-agent | govind@ | Approved | 2026-05-29 10:02 |
| CMP-2043 | mc/campaigns/2043 | New signups, week 1 | "Your first workspace, in plain English" | mei-marketing-agent | pending | Editing | — |
The row is the contract. The Mailchimp link is a pointer, not a copy. When the editor opens CMP-2043, the agent pulls the current segment count and recent open-rate trend from Mailchimp at read time, so the editor reviews against live numbers.
A worked workflow: trial day-7 nudge
A scheduled trigger in Mailchimp surfaces a segment of users at trial day seven. The agent reads the segment definition, fetches three recent product changes from the HubSpot product log, and drafts a subject line, preheader, and 90-word body. It writes a new row to Campaign Briefs with Status = Editing, the drafted copy, and a pointer to the Mailchimp campaign draft.
The editor opens Dock, reads the brief, and either approves or comments. On approval the row flips to Approved, a webhook tells Mailchimp to schedule the send, and the agent writes the final send time back to the row. Nothing leaves Dock without a named human on the row.
Why the split matters
Industry open rates hover around 35.6 percent and clicks near 2.6 percent according to Mailchimp's published benchmarks, and the Litmus State of Email Workflows report finds review and approval are the most frustrating bottleneck for marketing teams. Letting the agent draft inside Mailchimp removes the bottleneck only if you also keep an audit trail. Dock keeps it. Every approved subject line carries the editor's name, the agent's identity, and the timestamp, which is the basic shape covered in agent audit and compliance and agent identity.
This is the same pattern used across the Dock-for-marketing pillar, the Dock-for-sales outbound flows, and the Dock-for-ecommerce lifecycle email work. Rotate the agent on the schedule described in agent identity lifecycle and the attribution stays clean across the rotation.
Try Dock free with your Mailchimp workspace.
FAQ
Does the agent send the email directly? No. The agent drafts and queues. A named editor approves the Dock row, and Mailchimp executes the send under its own scheduled job.
What if the editor edits the copy in Mailchimp instead of Dock? The agent reads the Mailchimp campaign before send and writes the diff back to the brief row. The reviewed copy and the sent copy both live on the row.
Can the agent draft from HubSpot signals as well as Mailchimp segments? Yes. The brief row holds pointers to both platforms. The agent reads from whichever holds the current signal at draft time.
How is this different from an AI feature inside Mailchimp? Mailchimp's AI features draft in place. Dock adds the named editor row and the cross-platform pointer, so the approval is auditable outside the sending tool.
