What this means for your store
The native HubSpot ↔ GA4 link pipes website behaviour into CRM records and can send HubSpot lists back to GA4 audiences. When hubspotutk or a submitted email matches, page views, file downloads, and add_to_cart land on the contact timeline - not just in traffic reports. This is identity stitching, not server-side CAPI. It helps when sales asks “what did they browse before the demo?” and marketing needs one journey across blog, PDP, and quote form.
Scenario on a real storefront
A Shopify-plus-HubSpot kitchenware brand sees a repeat visitor download a sizing PDF, return via branded search, then submit a trade account form. After the integration, the contact record shows GA4 events beside CRM form activity:
# HubSpot → Settings → Integrations → Google Analytics 4
# Connect property → authorize Google account
# Map: page_view, form_submission, file_download, add_to_cart (custom as needed)
# Same hostname - HubSpot tracking + GA4 (GTM or gtag)
<script src="https://js.hs-scripts.com/PORTAL_ID.js" async></script>
# GA4 → Admin → Product links → HubSpot (audience export)
# Example: "MQL last 30 days" → GA4 remarketing (consent-checked)
# Weekly reconcile
# Contact timeline vs GA4 User Explorer - match hubspotutk + form email
# Cross-domain pitfall: Shopify checkout on shop.domain.com
# without linker → sessions never join to HubSpot contactIf checkout lives on a different subdomain than HubSpot landing pages, fix GA4 cross-domain linking first - otherwise PDP views stay orphaned from the CRM record.
What to do next
- Surface pre-form browsing on the contact card so reps see which SKUs were viewed before a wholesale quote - not only the thank-you page URL.
- Import HubSpot lifecycle segments to GA4 only after legal signs off; EEA/UK remarketing needs a documented lawful basis.
- Reconcile form_submission in GA4 against HubSpot form analytics weekly - drift shows up as CRM leads with no matching web event.
Bottom line
Connect GA4 to HubSpot, map commerce and lead events, and spot-check timelines against User Explorer before you blend acquisition reports with pipeline data.