Please do a GA4 report for Plunkett Homes last month.

Routes when a Jessie-assigned Asana task asks for GA4, Google Analytics, website analytics, traffic, engagement, landing-page, conversion-event, or channel-mix reporting.

• Live-data smoke on 2026-05-21 produced plunkett-homes-april-2026-internal-ga4-report.html from the hosted Asana path.
• Attachment verification confirmed live Google Analytics Data API evidence and no synthetic fixture caveat.
• Movement tables include movement percentages with positive/negative colour treatment for consultant scanning.
• The report model prioritizes digestibility by placing major wins and issues before detailed evidence.

---
name: ga4-reporting
description: Draft internal GA4 analytics reports from Asana task requests using live Google Analytics Data API evidence. Use when a task asks for a GA4, Google Analytics, website analytics, traffic, engagement, landing-page, conversion-event, or channel-mix report/review/summary.
---

# GA4 Reporting

## Quick Start

For Plunkett, a consultant can create an Asana task assigned to Jessie:

```text
Please do a GA4 report for Plunkett Homes last month.
```

Hosted Team Agent execution should infer:

- Skill: `ga4-reporting`
- Brand: `Plunkett Homes`
- Audience: `consultant`
- Reporting period: last complete calendar month
- Comparison period: prior calendar month

## Workflow

1. Confirm the task is agent-eligible and ready.
2. Fetch live GA4 evidence from the configured property.
3. Build a deterministic reporting model before any prose generation.
4. Put major wins, major issues, and watch items at the top for consultant scanning.
5. Follow with supporting detail for whole-site movement, channel mix, landing pages, conversion events, source/medium, devices, and locations.
6. Render an internal HTML report.
7. Attach the HTML report to the Asana task.
8. Post a review-required Asana comment.

## Evidence Rules

- Treat GA4 as the primary site analytics source, not as a replacement for ad-platform or Search Console evidence.
- Treat explanations as hypotheses, not attribution.
- Separate real performance movement from possible tracking, tagging, consent, attribution, or channel-grouping changes.
- State missing source caveats explicitly.
- Do not change GA4, GTM, website tracking, dashboards, or client-facing reports.

## Data Sources

Primary source:

- Google Analytics Data API for the configured GA4 property.

Core evidence:

- Overall sessions, users, engaged sessions, engagement rate, conversions, revenue, and event count.
- Default channel group movement.
- Source / medium movement.
- Landing-page session, conversion, and engagement movement.
- Conversion event and event-count movement.
- Device and location movement where useful.

## Approval Gates

Always gate:

- Client-facing action.
- External system write beyond the source Asana task comment/attachment.
- New secret or credential.
- GA4, GTM, website, dashboard, or tracking changes.

## Output Shape

Return:

- Major wins and issues first.
- Executive summary.
- Whole-site GA4 movement.
- Channel mix movement.
- Landing-page movement.
- Conversion and event movement.
- Source / medium movement.
- Device and location movement.
- Cause hypotheses with confidence, same-period or lagged timing, and follow-up checks.
- Data quality caveats.
- Recommended consultant actions.
- Approval-gated actions not taken.

• Ga4 Reporting Skill decisions/ga4-reporting-skill.md
• Ga4 Reporting Skill docs/runbooks/ga4-reporting-skill.md
• Team Agent Work Intake docs/runbooks/team-agent-work-intake.md