GA4 audience recipes by vertical

DTC ecommerce: the eight audience recipes

Audience	Definition	Activation
Cart abandoners (7d)	add_to_cart, not purchase, last 7 days	Retargeting on Meta, Google, email/SMS
High-intent prospects	begin_checkout, not purchase, last 14 days	Aggressive bid premium across paid channels
One-time purchasers	1 purchase, last 90 days	Second-purchase email flow, lookalike seed
Repeat customers	2+ purchases, last 180 days	VIP retention, advocacy campaigns
High-LTV customers	Lifetime revenue > $X threshold	Lookalike seed, exclusive offers
Lapsed customers	Past purchaser, no purchase 90+ days	Win-back email/SMS, retargeting
Likely 7-day purchasers	GA4 predictive audience	Bid amplification in paid campaigns
Browse abandoners	View product page 3+ times, no add to cart, last 7 days	Product-feature retargeting

B2B SaaS: the eight audience recipes

Audience	Definition	Activation
Pricing page viewers	page_view on /pricing, last 30 days, no signup	Demo request retargeting
Trial signups (engaged)	signup event + 3+ session events in trial period	Conversion-stage email and retargeting
Trial signups (stalled)	signup event + <2 session events	Activation email flow, customer success outreach
Demo requesters	demo_request event, last 60 days	SDR-led outreach + retargeting
Self-serve customers	subscription_start event with no sales touch	Onboarding flow, expansion retargeting
Enterprise customers	subscription_start with plan_tier = enterprise	Account-based expansion, executive content
Documentation deep readers	3+ docs page views, last 14 days	Power-user content, advanced feature retargeting
Churn-risk users	Active customer + GA4 churn probability > 70%	Customer success priority queue, retention offer

Lead generation: the six audience recipes

Audience	Definition	Activation
Form starters (incomplete)	form_start event, no form_submit, last 14 days	Retargeting with simplified form CTA
Phone callers (no form)	phone_click event, no form_submit, last 30 days	Form-completion retargeting, callback offers
Content downloaders	download_event, last 90 days	Nurture email + retargeting with next-stage content
High-engagement readers	3+ article views, 5+ min time on site, last 30 days	Lead capture retargeting
SQL audiences (CRM-uploaded)	From CRM where lifecycle_stage = SQL	SDR-aligned LinkedIn ABM + Google Ads
Opportunity stage (CRM-uploaded)	From CRM where opp_stage = Discovery+	Sales-cycle nurture, case-study retargeting

Local services: the six audience recipes

Audience	Definition	Activation
Service-page viewers (no contact)	page_view on service page, no phone_click or form_submit, last 7 days	Local Service Ads + Search retargeting
Quote requesters	quote_request event, last 30 days	Follow-up email/SMS, calendar booking
Phone callers	phone_click event	Source attribution + suppression from new-prospect campaigns
Past customers (job completed)	From CRM where job_status = completed	Review request, referral campaign, maintenance reminder
Service area prospects	Geographic targeting + service page view	Local Service Ads + Google Search ads + Local SEO
Emergency-keyword visitors	landing from emergency keyword ('plumber near me')	Top-of-page bid premium, 24/7 hours

Healthcare: the six audience recipes (HIPAA-aware)

Healthcare audiences must follow HIPAA and Google's healthcare advertising policies. Never include PHI (protected health information) in audience definitions or signals. Aggregated, anonymized behavioral signals only.

Audience	Definition	Activation
Appointment-page viewers	page_view on appointment booking page, no booking, last 7 days	Retargeting with appointment CTA (no condition-specific creative)
Provider directory researchers	3+ provider profile views, last 30 days	Generic provider/service retargeting
Insurance verifier users	insurance_check event	In-network promotion (general, not condition-specific)
Educational content readers	3+ educational article views, last 30 days	Brand-level retargeting, no condition-specific targeting
Past appointment attendees	From CRM where appointment_status = completed (de-identified)	Follow-up appointment reminder, review request
Patient portal active users	Patient portal login event	Engagement campaign, app download promotion

Media and content: the six audience recipes

Audience	Definition	Activation
Subscribers (active)	From CRM where subscription = active	Cross-sell, premium-tier upgrade, advocacy
Trial subscribers	trial_start event, last 30 days	Paid conversion campaigns, retention content
Lapsed subscribers	subscription = lapsed, last 60 days	Win-back offers, content highlights
Topic-specific readers	5+ pageviews in topic category, last 30 days	Topic-aligned subscription offers
Newsletter readers (engaged)	3+ email opens or clicks, last 30 days	Subscription conversion campaign
Paywalled-content abandoners	paywall_hit event, no subscription, last 14 days	Trial offer, single-article-purchase retargeting

The cross-vertical principles

1. Build by funnel stage, not just by event. A "viewed pricing" audience is more useful split into "viewed pricing but no demo" vs "viewed pricing and requested demo."
2. Include exclusions in every audience. Cart abandoners minus customers = the right retargeting audience. Without exclusions, you retarget your existing customers with prospecting offers.
3. Match audience definitions to bid strategy. High-LTV audiences deserve aggressive bidding. Cold audiences deserve testing budgets.
4. Tier audiences by intent. Just like keywords (see search intent tiering), audiences should be tiered by behavioral intent.
5. Activate everywhere, not just Google. The same audience that bids high in Google Ads should bid high in Meta, TikTok, and email. Cross-channel consistency compounds.

Are these recipes a starting point or final state?

Starting point. Audience design is iterative. Build the basics first, measure activation lift, then refine based on what actually moves the business.

How do I avoid HIPAA violations in healthcare audiences?

Never include PHI (protected health information) in audience definitions or signal events. No condition-specific targeting, no patient identifiers. Aggregated behavioral data only. When in doubt, consult legal counsel — penalties are significant.

Can I run lookalike audiences off these?

Yes — high-LTV customers and converted users are the most useful lookalike seeds. Build them in GA4, export to each platform's lookalike feature (Google Similar Audiences, Meta Lookalike, etc.).

What's the right audience refresh cadence?

Behavioral audiences refresh automatically as new events come in. Predictive audiences refresh as the model retrains. CRM-imported audiences refresh per your CRM-to-platform sync cadence — daily or hourly is standard.

How many audiences should I have?

15-30 for a mature account is typical. Below 10, you're missing leverage. Above 50, audience sprawl makes activation hard to manage. Audit quarterly and prune low-usage audiences.

Where do I track audience performance?

In each downstream platform's reporting (Google Ads campaign performance by audience, Meta Ads ad-set-level audience performance). GA4 itself doesn't directly report "this audience converted at X rate" — that's downstream.

Operating checklist

1. Define the business outcome the audience supports before building.
2. Confirm GA4 has the events and parameters the audience condition needs.
3. Build the audience in GA4 with clear naming conventions.
4. Link GA4 to Google Ads, Meta Ads, and any DSPs that will activate it.
5. Verify audience populates within 24-48 hours.
6. Set a refresh cadence and ownership for the audience.
7. Document the audience purpose, definition, and downstream usage.