MARKETING

UTM taxonomy governance agent with vocabulary proposals

An agent reviews quarantined links weekly, distinguishes genuine errors from legitimate new values marketers actually need.

CategoryMarketing
Enginepaperclip
Difficultyadvanced
Triggerschedule
Steps6
Setup~25 min

How it runs

The automated pipeline, trigger to output.

  • TriggerWeekly governance schedule fires
  • ActionPull quarantine backlog from BigQueryGoogle BigQueryBigQuery
  • LogicCluster violations and classify error vs legitimate new value
  • ActionDraft proposed taxonomy additions with rationale
  • ActionPost proposals to Slack for approvalSlack
  • OutputCommit approved vocabulary to taxonomy store in NotionNotionNotion

What it does

This workflow runs an agent that governs the taxonomy itself, not just enforces it. Weekly it reviews the backlog of quarantined links, clusters the recurring violations, and reasons about which are true mistakes versus which represent a legitimate new source, medium, or campaign pattern the vocabulary should adopt. It drafts proposed taxonomy changes with rationale and routes them for human approval before anything is committed.

When to use it

Use it when a static taxonomy goes stale — new channels, new partners, new naming conventions keep getting quarantined and someone has to decide whether the rule or the link is wrong. This keeps the vocabulary current without letting it sprawl.

How it works

  1. 1A weekly schedule starts the governance run.
  2. 2The agent pulls the recent quarantine backlog from BigQuery.
  3. 3It clusters violations and reasons about error versus legitimate new value.
  4. 4It drafts proposed taxonomy additions with rationale and impact.
  5. 5Proposals are posted to Slack for Chairman approval.
  6. 6On approval, the agent commits the updated vocabulary to the taxonomy store.

Set it up

What you configure once, before turning it on.

  1. 1
    Connect BigQueryDatasets, queries, schemas.
  2. 2
    Connect SlackChannels, DMs, threads, mentions.
  3. 3
    Connect NotionPages, databases, comments.
  4. 4
    Set each agent's modelWe leave models unset so you pick the tier — fast + cheap, or top-quality.
  5. 5
    Tune it to your dataEdit the prompts, filters, and field mappings so it matches how your team works.
  6. 6
    Test, then turn it onRun once against a sample, confirm the output, then enable the trigger.

Run this workflow in your colony.

14-day trial. No DevOps. No Sales call. Provisioned in under a minute.