The short answer
A marketing agent can draft a HubSpot campaign in seconds. Brand sign-off is the slow part, and it is the part nobody can reconstruct three months later. Dock fixes the reconstruction problem. The agent writes the brief into a Dock table, a human brand reviewer approves or rejects each section, and the row carries the agent identity, the decision, and the timestamp. The HubSpot campaign and the Iterable journey stay where they live. The interpretation, the rationale, and the sign-off live in Dock for marketing.
The architecture
HubSpot Marketing Hub and Iterable 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.
This matters because the HubSpot campaign object has a status field, not a reasoning field. The Iterable journey has a send-time, not a rationale. The campaign brief, the segment logic, the subject-line variants, the brand voice check, and the approval chain all need to be reconstructable on demand. They cannot live as ephemeral chat output.
One Dock surface: the Campaign Brief table
| Brief ID | HubSpot Campaign | Segment | Drafted by | Brand reviewer | Status | Sign-off at |
|---|---|---|---|---|---|---|
| CB-218 | hubspot://campaign/91044 | NA mid-market, opened in 30d | agent:mei-marketing-v3 | priya.s@ | Approved with edits | 2026-05-28 14:02 |
| CB-219 | hubspot://campaign/91051 | Iterable journey J-7782 reactivation | agent:mei-marketing-v3 | dan.t@ | Rejected, off-voice | 2026-05-28 16:40 |
| CB-220 | hubspot://campaign/91063 | Enterprise expansion, ABM tier 1 | agent:mei-marketing-v3 | priya.s@ | Pending | — |
Each row links out to the HubSpot campaign and the Iterable journey. Each row also links inward to the prompt the agent used, the source positioning doc, and the brand-voice rubric the reviewer scored against. The row is the durable artifact. The campaign object is the live system.
One worked workflow
A reactivation campaign starts when Iterable flags 4,200 dormant accounts. The agent reads the segment, pulls the last three winning subject lines from HubSpot, drafts a brief with two subject variants, body copy, and a sending window. It writes the brief to row CB-219 and pings the brand reviewer in Dock.
The reviewer rejects the brief. The voice reads off-brand for the reactivation tier. The reviewer leaves a comment, marks the row Rejected, and the agent re-drafts using the comment as input. Round two passes. The agent flips the HubSpot campaign from Draft to Scheduled, writes the send-time back to the row, and the audit trail is closed. A Q3 brand audit can pull every reactivation brief by agent, by reviewer, by decision, in one query.
Why it matters
HubSpot's 2026 State of Marketing report finds 80% of marketers use AI for content creation and 75% for media production, with 61% calling AI the biggest disruption in 20 years (HubSpot, 2026). At the same time, Gartner's 2025 CMO Spend Survey shows marketing budgets flatlined at 7.7% of revenue with 59% of CMOs reporting insufficient budget to execute strategy (Gartner, 2025). The pressure is to ship more agent-drafted work with the same brand bar. That requires agent identity on every draft and a reviewable surface for brand sign-off.
CTA
Start a Campaign Brief table in your Dock workspace this week. One row per HubSpot campaign, one column for the agent that drafted it, one column for the brand reviewer who signed.
FAQ
Does Dock replace HubSpot Marketing Hub? No. HubSpot remains the campaign system of record. Dock holds the brief, the agent identity, and the brand decision. The same pattern works for sales handoffs in Dock for sales.
How does the agent prove it drafted a specific brief? Every row carries the agent identity at write time. The identity is issued through the agent identity lifecycle and verified against the workspace registry, not the human credential the agent borrowed.
Can brand reviewers reject parts of a brief? Yes. The reviewer marks the row Rejected with a section-level comment. The agent re-drafts only the rejected sections and writes a new row version. The history is preserved for agent audit and compliance.
What if the same agent handles SEO content? The same identity pattern applies. The SEO brief table lives next to the campaign brief table, with the same reviewer column. See Dock for SEO for the content-side variant.
