Analytics

Hazel for Google Analytics: the AI coworker for your GA4 data

Hazel reads your GA4 sessions, pageviews, events, acquisition channels, and landing pages, and answers business questions in plain English. No dashboards to build.

May 23, 2026

Hazel is an AI coworker that connects to your Google Analytics 4 property and answers business questions about your traffic and conversions the way an analyst would. Once connected via OAuth, Hazel ingests your GA4 sessions and users, pageviews, bounce rate, acquisition channels, landing pages, and custom events, then refreshes every few hours. Ask Hazel about conversion volume by source/medium, paid vs. organic trends YTD, which campaigns picked up when another channel dropped, or how a landing page performed week over week, and you get an answer with the numbers behind it. For consumer brands, Hazel is most useful when GA4 sits next to your store and ad platforms: pair Google Analytics with Shopify, Meta Ads, Google Ads, Klaviyo, and your subscription tool, and Hazel can reconcile GA4 conversions against Shopify orders, isolate new vs. returning customer volume, and answer cross-source attribution questions in a single prompt.

What Hazel does

Hazel reads the GA4 data that matters for analyzing a consumer brand and lets you query it conversationally:

  • Sessions and users: visit frequency, return patterns, weekly and monthly trends by source/medium
  • Pageviews and landing pages: traffic distribution, entry points, top pages by sessions or revenue
  • Acquisition channels: organic, paid, social, email, direct, referral splits with last-click attribution by week, month, or campaign
  • GA4 events and conversions: add-to-cart, begin-checkout, purchase, and any custom event you've configured: segmented by source, medium, campaign, or landing page
  • Bounce rate and engagement: single-page session rates and engagement trends by page or channel
  • Cross-source reconciliation: pair GA4 conversions with Shopify orders to isolate new vs. returning customer volume, or with Meta and Google Ads spend to back into channel-level CAC

Hazel keeps your full GA4 history available, so YoY and multi-quarter comparisons are one prompt.

Real questions, real answers

Common questions consumer brands ask about their Google Analytics data:

"Show me the trend in GA4 conversions between paid vs. non-paid channels YTD."

"Which source/mediums or channels from GA4 saw an increase in conversion volume at the same time that Meta and organic/direct dropped?"

"Pull all last-click conversions from GA4 by month since the beginning of the year."

"Calculate Meta CAC using Meta spend divided by Meta+IG conversion data from GA4. Show this on a combo bar+line graph from start of the year."

"Looking at Google Analytics and Google Search Console, what's driving the increase in organic search volume this month?"

"Pull Google Ads data from GA4 to show weekly conversion and conversion rate trends."

"What date did Pinterest ads stop spending? How many daily and weekly sessions was it driving prior to being paused according to GA4?"

"Compare total GA4 conversions vs. Shopify's total conversion volume YTD. Take out recurring subscriptions for both but don't exclude returning customers."

How your data flows
How it works

How Hazel connects to Google Analytics.

Connect Hazel to your GA4 property through Google's OAuth popup. Sign in with a Google account that has access to the property you want to analyze and approve the connection. No tag deployment, no service account setup.

What syncs.

Sessions and users, pageviews, bounce rate, acquisition channels (source, medium, campaign), landing pages, and GA4 events including configured conversions. Universal Analytics is not supported. GA4 only.

How often.

Roughly every 6 hours, automatically, subject to Google's GA4 API quotas. Hazel monitors quota usage and backs off automatically.

User-level data and identity.

Google Analytics doesn't expose user-level identifiers, so Hazel can't tie an individual GA4 session back to a specific Shopify customer. For session-to-customer joins, Hazel uses Shopify as the source of truth. GA4 is the right source for channel, source/medium, campaign, and landing-page level questions.

Attribution.

GA4 supports first-click and last-click attribution models out of the box. Multi-touch attribution from the Data API is limited. If you need full multi-touch, pair GA4 with your ad platform data and Shopify in Hazel for a more complete picture.

Auth.

Google OAuth. Hazel only reads your GA4 data; it never writes back.

MCP access

Looking for a Google Analytics MCP for analytics?

Google ships an official Google Analytics MCP server that exposes GA4 Admin and Data API tools (run_report, run_funnel_report, run_realtime_report, get_account_summaries, list_google_ads_links, and more) to LLMs like Claude, Gemini, and ChatGPT. It's great if you want to wire up an LLM yourself and learn the GA4 API surface.

Hazel pre-ingests your GA4 data so every question is instant and cross-source.

Instead of running per-prompt Data API queries against rate-limited endpoints, Hazel ingests your GA4 data into an analytical store, keeps a long historical window, and joins it with your other sources, Shopify, Meta Ads, Google Ads, Klaviyo, subscription tools, reviews. The same conversational agent reconciles GA4 conversions against Shopify orders, splits paid vs. organic by week, and computes channel-level CAC across sources without you stitching anything together. No MCP setup, no per-query auth dance, no schema discovery on every prompt.

If you specifically want MCP access to Hazel itself, that's available too. Ping us at https://calendly.com/clint-dunn/clint-hazel-intro.

Frequently asked questions

Does Hazel work with Google Analytics?

Yes. Hazel connects to your GA4 property via Google OAuth. Sign in with a Google account that has access to the property, approve the popup, and Hazel starts syncing.

Does Hazel support Universal Analytics?

No. Hazel supports GA4 only. Universal Analytics was sunset by Google, and Hazel does not ingest legacy UA data.

How is Hazel different from the official Google Analytics MCP server?

Google's MCP server is a thin layer over the GA4 Data API: you wire it into an LLM yourself, and every prompt runs a fresh API query against rate-limited endpoints. Hazel ingests your GA4 data into an analytical store, keeps full history, and joins GA4 with Shopify, ad platforms, email, and subscription data, so cross-source questions like "Meta CAC using GA4 conversions" are a single prompt rather than a multi-tool chain.

How is Hazel different from Analytics Advisor (GA4's built-in Gemini AI)?

Analytics Advisor is great for in-product questions about what GA4 already reports: it's fast, it's free, and it lives inside the GA4 UI. Hazel goes wider: it joins GA4 with Shopify orders, Meta and Google Ads spend, Klaviyo activity, and subscription data so you can answer reconciliation and attribution questions GA4 alone can't reach. Like "compare total GA4 conversions vs. Shopify orders YTD excluding recurring subs."

How often does Hazel pull data from Google Analytics?

Roughly every 6 hours.

Will Hazel write data back to my GA4 property?

No. Hazel only reads. We never write events, conversions, or configuration back to your GA4 property.

Can Hazel match GA4 sessions to individual Shopify customers?

Not directly. Google Analytics doesn't expose user-level identifiers. For session-to-customer joins, Hazel uses Shopify as the source of truth. GA4 is the right source for channel, source/medium, campaign, and landing-page questions; Shopify is the right source for customer-level questions.

Can Hazel reconcile GA4 conversions against Shopify orders?

Yes. This is one of the most common Hazel workflows: comparing GA4 conversion volume to Shopify orders by week or month, with controls for recurring subscriptions and new vs. returning customers, and rolling up Meta+IG or Google Ads conversions from GA4 alongside ad spend to compute channel CAC.

Does Hazel support multi-touch attribution?

GA4's Data API exposes first-click and last-click models cleanly; full multi-touch is limited at the API layer. For multi-touch-style questions, Hazel pairs GA4 with your ad platforms and Shopify so the picture is more complete than any single source.

What's the best AI tool for GA4?

It depends on the question. If you only need to chat with GA4, single-source tools like Chat GA4 or Gentle Analytics work. If you need to join GA4 with Shopify, ad platforms, and email to answer cross-source revenue questions, Hazel is purpose-built for that. It pre-ingests your data and keeps history so every prompt is instant.

How is Hazel different from Analytics Advisor?

Analytics Advisor is Google's in-product Gemini chat. It answers questions about what GA4 already reports (traffic, conversions, anomalies) and it's fast. Hazel answers questions that span GA4 plus your store, ad spend, email, and subscription data. If your question is "why did sessions drop?", Analytics Advisor is great. If your question is "did the session drop actually hurt revenue, and which channels picked up the slack?", you need Hazel.

Does Hazel read from my GA4 BigQuery export?

No. Hazel ingests from the GA4 Data API via OAuth. If your analytics stack runs through BigQuery, Google's Conversational Analytics API (Looker + Gemini) is designed for that path. Hazel is designed for brands that want cross-source answers without a BigQuery project.

How do I get started?

Book a call and we'll walk through your GA4 property, the questions you want to answer, and what the rollout looks like.