GA4 Analyst — Claude AI Skill for Google Analytics 4 Insights

Claude can’t see your analytics — gives generic advice like “check your bounce rate” without knowing what your bounce rate actually is

Navigates your live GA4 account directly — reads your actual sessions, channels, conversion rates, and top pages before saying anything

You manually pull reports, copy numbers into chat, and hope Claude interprets them correctly — error-prone and time-consuming

No copy-pasting — the skill navigates to the right report, takes a screenshot, and uses page extraction to reliably capture the numbers

“Why did traffic drop?” gets a list of possible causes — algorithm update, seasonality, tracking issue — with no way to check which one applies to you

Pulls your actual channel breakdown before and after the drop, checks the traffic acquisition report, and names the specific source that changed

Analysis is limited to whatever you screenshot and paste — if you missed a relevant report, that insight is lost

Knows exactly which GA4 reports answer which questions — navigates up to 15 report types including Explore, Funnels, Path Analysis, and Search Console

No export — insights live in the chat and disappear when the conversation ends

Exports a full analytics report as a shareable Markdown summary or structured .xlsx spreadsheet with tabs for Traffic, Conversions, Audience, and Recommendations

Data quality issues go unnoticed — (not set) sources, sampling warnings, and tracking gaps never get flagged

Flags data quality issues before drawing conclusions — (not set) in source/medium, GA4 sampling warnings, missing conversions, and tracking gaps all called out

Opens Google Analytics, checks which account and property is loaded, and confirms the date range before touching any data. If there are multiple properties, asks which one to use. Never analyzes the wrong account.

“Why is traffic dropping?” → Traffic Acquisition report. “Which pages convert best?” → Landing Pages + Conversions. “Where are users dropping off?” → Funnel Exploration in Explore. The skill knows which of the 15+ GA4 report types answers each question — no manual digging.

Question type → GA4 location
"Why is traffic dropping?"    → Reports → Acquisition → Traffic acquisition
"What are my top pages?"      → Reports → Engagement → Pages and screens
"How are conversions?"        → Reports → Engagement → Conversions
"Where do users drop off?"    → Explore → Funnel exploration
"What does my audience look like?" → Reports → User → User attributes

Takes a screenshot and uses page text extraction to reliably capture metric values, table rows, and numbers — not just what’s visible on screen. If data is paginated, scrolls to get the full picture. If a comparison period is needed, enables it in the date range picker.

Traffic trends and channel share, content engagement and problem pages, conversion rates by channel, audience demographics, and retention cohorts — all covered in one analysis pass.

Every analysis follows the same structure: Key Metrics, What the Data Shows, Standout Findings, Ranked Recommendations, and a Watch List for anomalies and data quality issues.

On request, pulls 6–12 months of historical data and builds three scenarios — optimistic (current growth continues), base (seasonal patterns applied), and risk (declining trend unaddressed).

Saves a Markdown summary or builds a structured .xlsx spreadsheet — Traffic, Conversions, Audience, and Recommendations tabs — ready to share with stakeholders or your team.