I Have Not Manually Updated a Campaign Task in Months. Here Is the Agent Running Every Brand in Notion.

Campaign management is the invisible tax on every DTC operator's week. Not the strategy, not the creative, not the actual execution: just the coordination layer. The status checks, the "where are we on the email brief" messages, the task updates nobody made, the overdue deliverables nobody caught until the campaign already went live with something missing.

Running multiple brands means running multiple Notion workspaces. Each with its own campaign calendar, live flows, creative briefs, and approval chains. The information exists. It just does not surface itself. Before this agent ran, I was spending the equivalent of a full working day per week across client workspaces, not doing work but tracking whether work had been done. Every week spent coordinating is a week not operating.

What the agent actually does

The agent runs on two cadences: a lightweight daily scan at 7am, and a full weekly review every Monday morning. Here is the complete workflow.

The context layer: why it feels like an ops manager

The difference between a useful agent and a noisy one is context. The agent does not treat all tasks equally because not all tasks are equal. At the start of each run, it reads a client context brief containing the campaign calendar for each brand, the approval chain, which tasks are on the critical path, and the naming conventions used in that workspace. This brief is written once per client during onboarding and updated when the campaign structure changes.

That context layer prevents false positives. A task that looks overdue might be correctly postponed because a brief above it shifted. An unowned task might be intentionally open pending an external deliverable. Without context, the agent alerts on everything and becomes background noise within a week. With context, it alerts only on things that actually need attention. That is the distinction operators feel immediately: the first time the agent surfaces something real that would have slipped through, with nothing else alongside it.

What the Monday digest looks like

Below is a simplified version of the iMessage that arrives every Monday morning. Brand names are removed but the structure is live.

The full Notion Campaign Health page is linked in each message. Each brand section links directly to the relevant task database. The whole review takes under five minutes. The alternative, which involved manually checking each workspace, chasing owners, and reconciling campaign dates against deliverable statuses, used to consume most of a Monday morning.

Inside the System

How we build this for brands

For the brands we work with, the Notion campaign agent sits inside a broader operations layer that connects every channel into one coherent view. When a Klaviyo flow is deployed, the corresponding Notion deliverable task updates to complete automatically. When a Meta creative is rejected during review, a blocker task is created and assigned before anyone has to manually log it. The campaign tracker is not reading a static database. It is reading a workspace that itself updates in response to what is actually happening across channels.

The result for operators is a single source of truth across every channel without the manual overhead of maintaining it. Campaign status reflects reality, not the last time someone remembered to update a task. Part of this runs live for portfolio brands today; the full system is what we deploy when we take a brand on.

Frequently asked questions