UTM and campaign naming hygiene check
A weekly check of campaign links and analytics data that finds inconsistent UTM tags and campaign names, groups likely duplicates, and suggests corrections for the marketer to apply.
The use case
Marketing reports become unreliable when the same campaign is tagged as facebook, Facebook, and fb, or when links go out without UTM parameters at all. Small teams notice only when a report shows dozens of tiny sources that should be one. The agent reads traffic source data from Google Analytics and the team's link-building sheet each week and compares tags with the written naming convention. It groups variants that likely mean the same thing, lists links missing tags, and flags new campaign names that do not follow the convention. It suggests corrected tags and a mapping for past data. The marketer reviews the suggestions, fixes links going forward, and decides how to treat past data in reports. The agent does not edit live ads, emails, or analytics settings.
Example workflow
- Each week, the agent pulls source, medium, and campaign values from Google Analytics for the past week.
- It reads the team's UTM link sheet and the written naming convention stored in Notion.
- It groups variants that likely refer to the same source or campaign and flags values that break the convention.
- It lists recently shared links with missing or broken tags, where link data is available.
- A short report with suggested corrections and a mapping table is posted to Slack for the marketer.
- The marketer updates links and templates, and approved mappings are added to reporting in Looker Studio.
Tools
- Google Analytics
- Google Sheets
- Looker Studio
- Notion
- Slack
- Claude
Risks and guardrails
- Variants are grouped that are actually different campaigns — guardrail: groupings are suggestions, and the marketer approves each mapping.
- Past reports change without anyone noticing — guardrail: mappings are applied in a separate reporting layer with a change log.
- The naming convention is unclear, so suggestions conflict — guardrail: the agent flags gaps in the convention for the marketer to settle.
- The agent edits live campaigns — guardrail: it has read-only analytics access and no access to ad or email platforms.
Related recipe
Weekly KPI Digest from Google Sheets with Source-Cited Numbers: An agent reads your defined metrics from Google Sheets, compares them to last week and targets, and drafts a plain-English digest where every number cites its cell. The owner approves before it goes to Slack or email.
Want it set up for you? Request a quote.
Sources
Written by Agents Lookup. Credits and official pages for the tools this use case names:
- Google Analytics MCP server (official page): github.com/googleanalytics/google-analytics-mcp (checked 2026-11-09)
- Notion MCP server (official page): developers.notion.com/guides/mcp/overview (checked 2026-11-09)
- Slack MCP server (official page): docs.slack.dev/ai/slack-mcp-server (checked 2026-11-09)
- Claude (official page): claude.ai (checked 2026-11-09)