GA4 Audiences Guide: Build Segments That Drive SA Conversions

A ga4 audiences guide gives South African marketers the blueprint to stop treating analytics as a passive reporting layer and start treating it as a live segmentation engine. Used correctly, conversion rate optimisation in South Africa depends as much on who you're targeting as on what your pages say — and GA4 audiences are the mechanism that connects behavioural data to action.

Build the right segments, link them to Google Ads, and you're not waiting for visitors to return by chance: you're surfacing your brand to the exact users most likely to convert, based on what they already did on your site.

South Africa's ecommerce user base is forecast to grow from approximately 11.7 million in 2025 to 21.5 million by 2029 (Statista) — a market expanding fast enough that the difference between businesses running targeted audience strategies and those running broad traffic campaigns will widen each year. Most SA businesses running Google Analytics 4 are using it passively — dashboards, session counts, traffic sources. The audience-building layer sits untouched in Admin, and with it goes the ability to re-engage cart abandoners, suppress recent buyers from prospecting campaigns, and identify your highest-value spenders before they defect. This guide covers every audience type GA4 offers, how to create and configure them correctly, how to export them to Google Ads, and what POPIA requires before you activate any of it.

What Are GA4 Audiences and How Do They Differ from Segments?

A GA4 audience is a named group of users who share defined characteristics — behavioural, demographic, or predictive — that GA4 monitors in real time and can share with linked advertising platforms. Unlike segments, which are temporary filters applied to your reports for one-off analysis, audiences are persistent: once created, GA4 continuously evaluates every new user session against the audience's conditions and adds qualifying users automatically.

The practical difference matters for CRO work. A segment tells you what happened to a group of users in a past period. An audience tells Google Ads which users to show your campaigns to right now. Both are useful; only audiences drive action outside of GA4 itself.

The non-retroactive nature of audiences is the rule most SA businesses learn the hard way. You cannot go back and populate an audience with historical users — the moment you create it is the moment it starts counting. Create your core audiences now, even if you're not running remarketing campaigns yet. The data will be ready when you are. One of the most consistent pieces of advice in any ga4 audiences guide is simply this: build first, campaign second.

The Four Types of GA4 Audiences Explained

GA4 organises its audience options into four distinct types, each suited to a different level of setup effort and data availability.

Predefined audiences ship with every GA4 property: All Users and Purchasers. These are the baseline lists — useful for broad remarketing but not segmented enough for CRO work. Start here, but don't stop here.

Suggested audiences (also called template audiences) are GA4's pre-built options covering about 18 common use cases. GA4 identifies which types are most relevant for your property. Cart abandoners is the most commercially important — the global average cart abandonment rate sits at 70.22% according to Baymard Institute's analysis of 50+ studies, making a cart abandoner audience typically your single largest pool of warm, high-intent users.

Cart abandonment in South Africa follows similar patterns, compounded by payment gateway friction and other local market factors that interrupt the checkout flow. This is where a suggested audience template earns its place immediately, with no custom configuration required.

Custom audiences give you full control. You define conditions based on events, event parameters, user properties, demographics, or sequences of behaviour. A custom audience might be "users who viewed a product page more than twice but did not add to cart" or "users who spent more than three minutes on the pricing page." You can stack up to 10 conditions per audience and add exclusion rules to remove existing customers.

This is where landing page CRO meets audience precision — the more specific your audience, the more targeted your message can be.

Predictive audiences use machine learning and are covered in detail in their own section below.

GA4 Audiences Guide: How to Create an Audience Step-by-Step

Creating a GA4 audience takes less than five minutes once your events are tracking correctly. The path is the same for all audience types.

1. Navigate to Admin — click the gear icon at the bottom left of GA4.
2. Open Audiences — under the Property column, click "Audiences".
3. Click "New audience" — you'll see three options: start from scratch, use a suggested template, or use a predictive template.
4. Name it clearly — use a convention you'll recognise at a glance inside Google Ads. "Cart Abandoners — 30 Day" is better than "Audience 1".
5. Add conditions — choose from events, parameters, demographics, or behavioural triggers. Sequences (user did X then Y) and exclusions are available. Maximum 10 conditions.
6. Set membership duration — the default is 30 days. For cart abandoners this often works well. For high-value customer audiences you may want up to 180 or 540 days. The duration resets each time a user re-qualifies.
7. Save — the audience becomes active immediately and begins populating. Expect 1–2 days before users appear.

Exporting GA4 Audiences to Google Ads: The Activation Step

A GA4 audience that stays inside GA4 is analytical intelligence with no output. The conversion layer happens when you export audiences to Google Ads and use them to shape bidding, targeting, and creative delivery.

The setup requires three things to be in place simultaneously:

1. Link GA4 to Google Ads — Admin > Property > Product Links > Google Ads Links. Both accounts must be managed by the same individual or an authorised user of both.
2. Enable Google Signals — Admin > Data Collection > Google Signals. This unlocks cross-device tracking and is required for audiences to populate in Google Ads.
3. Keep Personalised Advertising active — this setting must remain enabled in the Google Ads link for audiences to flow through.

Once linked, your GA4 audiences appear automatically in Google Ads' shared library. They populate with up to 30 days of historical audience data from the day the link is established.

If your GA4 audience shows 800 users but Google Ads shows zero or "audience too small", check which campaign type you're trying to apply it to. Display requires 100; Search requires 1,000. For smaller SA ecommerce sites, Display and YouTube remarketing will unlock first. A/B testing your remarketing creative against these segments — cart abandoner creative vs. product-page-browser creative — is where the CRO value compounds.

Create your core audiences today. Run the campaigns once the lists reach the minimum threshold — the audience data is accumulating whether or not a campaign is active against it.

Predictive Audiences: Forecast Behaviour Before It Happens

Predictive audiences use GA4's machine-learning models to group users by forecast behaviour — purchase probability, churn probability, and predicted revenue — over the next 7 to 28 days, without manual condition-setting. Any complete ga4 audiences guide has to cover this layer specifically, because it's where analytics stops being a backward-looking dashboard and becomes a forward-looking decision engine that turns reactive remarketing into proactive revenue protection.

GA4 offers three underlying predictive metrics:

From these three metrics, GA4 generates five suggested predictive audience templates:

• Likely 7-day purchasers — purchase probability above the 90th percentile. Your hottest leads.
• Likely first-time 7-day purchasers — same threshold, but only users with zero purchase history. New customer acquisition target.
• Likely 7-day churning purchasers — past buyers with churn probability above the 80th percentile. The win-back opportunity.
• Likely 7-day churning users — all users (not just buyers) with high churn probability. Useful for engagement campaigns.
• Predicted 28-day top spenders — predicted revenue above the 95th percentile. Your VIP tier — the users worth showing higher-margin products or exclusive offers to.

For SA businesses with seasonally concentrated sales (Black Friday, month-end pay cycles), predictive audiences are particularly valuable. A "Likely 7-day purchasers" audience captured before a peak period lets you concentrate paid media spend on users who are already on the verge of buying — rather than broad prospecting into cold traffic. Pair these audiences with checkout page improvements and the combination compounds: you're bringing more ready-to-buy users to a page that converts more of them.

POPIA Compliance for Behavioural Tracking in South Africa

GA4 audiences are built from behavioural tracking data — and in South Africa, that data cannot be collected without explicit opt-in consent under the Protection of Personal Information Act (POPIA). This is not optional, and it is not the same as the European GDPR framework: POPIA mandates a genuine opt-in model, not a passive "by continuing to browse" implied consent.

What this means for your GA4 audience strategy:

• Your consent banner must require active acceptance. Pre-ticked boxes, implied consent, and banners that only show an "Accept" button are non-compliant. The reject or "Manage Preferences" option must be as prominent as Accept.
• Analytics cookies are non-essential. The tracking that powers GA4 audiences — persistent identifiers, device fingerprinting, behavioural session data — requires consent before it fires. You cannot assume legitimate interest applies to behavioural advertising data.
• Consent expires after 12 months. Your consent management platform must re-request consent annually and maintain timestamped records of each user's consent choices.
• GA4 data retention should be set deliberately. Inside GA4 Admin > Data Settings > Data Retention, review your user-data retention period and align it with your POPIA data minimisation obligations.

The practical implication for audience size: a site that collects clean opt-in consent will see a portion of users accepting analytics cookies — the precise share depends on banner design and placement. Your GA4 audiences will reflect consented users only, not total sessions. This is the accurate, legally defensible number, and this ga4 audiences guide treats that distinction as a feature rather than a limitation. Run your conversion rate benchmarks against consented-traffic GA4 data, not against total server-side session counts, or your comparisons will be inflated.

See our dedicated post on cookie consent and analytics for South African businesses for implementation detail on consent banners and GA4 configuration.

Why South African Businesses Choose Growth Pulse Media for GA4 and CRO

A GA4 audience setup is ready to drive revenue when three things align: the correct events are firing (purchase, add-to-cart, form-submit), the audiences are actively syncing to Google Ads, and the campaign structure is built to act on the intelligence those audiences provide. Most SA properties we review have one of the three — rarely all three connected. Our conversion rate optimisation service is built to close that gap: events, audiences, campaigns, and the CRO layer working together rather than in isolation.

Most digital agencies in South Africa either run ads or run analytics — few do both at the level where GA4 audience strategy connects directly to campaign outcomes. Dirk founded GPM after running South African ecommerce operations directly — managing Shopify stores, working with PayFast and Peach Payments integrations, and learning where SA customers actually drop off. That operational background means our GA4 audience builds reflect real checkout flows and real SA consumer behaviour, not global templates applied without local context.

We work with a deliberately limited client load so that every GA4 property gets senior attention — not a junior analyst running someone else's playbook. Our stack: GA4, Google Ads, and Looker Studio, configured end-to-end so audiences feed campaigns, suppression rules exclude existing buyers, and predictive lists inform bid strategy rather than sitting idle.

Who This Is NOT For

Frequently Asked Questions About GA4 Audiences

What is a GA4 audience and how is it different from a segment?

A GA4 audience is a persistent group of users who match defined conditions, monitored in real time by GA4 and shareable with linked advertising platforms like Google Ads. A segment is a temporary filter applied to your reports for analysis only — it doesn't persist between sessions or get exported to ad platforms. Audiences are for activation; segments are for investigation.

How many audiences can I create in GA4?

Standard GA4 properties support up to 100 audiences. GA4 360 properties allow up to 400. Once you reach the limit, you must archive an existing audience to create a new one. Archived audiences stop collecting new users but retain their historical data.

How long does it take for a GA4 audience to populate in Google Ads?

After creating an audience and linking GA4 to Google Ads, expect 1–2 days before users start appearing. The audience will also be backfilled with up to 30 days of historical data from users who already qualified at the time of linking. You'll only be able to use the audience in campaigns once it reaches Google Ads' minimum: 100 users for Display and YouTube, 1,000 for Search.

What are predictive audiences in GA4 and how do I enable them?

Predictive audiences use machine learning to group users by predicted future behaviour — purchase probability, churn probability, and predicted revenue are the three underlying metrics. They're not manually enabled: they become available when your GA4 property meets Google's data thresholds for the predictive model. Correctly configured purchase event tracking is the practical prerequisite. If predictive templates are greyed out in your audience builder, your property hasn't yet qualified.

Does POPIA affect my ability to use GA4 audiences in South Africa?

Yes. POPIA requires explicit opt-in consent before you deploy analytics cookies that feed GA4's behavioural tracking. Without a compliant consent mechanism, the data underpinning your audiences is collected without a lawful basis. Your audiences will only include users who consented to analytics tracking — which is the correct and legally defensible position. A well-implemented consent banner reduces audience size relative to raw traffic, but makes the data trustworthy and compliant.

Can I use GA4 audiences for CRO analysis without Google Ads?

Yes — audiences also work as comparison groups within GA4's standard reports and Explorations. You can compare conversion rates between audience members and non-members, analyse path differences between high-intent and low-intent users, and use audience membership as a dimension in your reports. This analysis layer exists entirely within GA4 and doesn't require a Google Ads link. The Google Ads export adds the paid activation layer on top of the analytical value.

Dirk van Greuning Founder, Growth Pulse Media

Founder of Growth Pulse Media and a specialist in South African search dominance. Dirk translates his experience in scaling South African businesses into high-velocity digital strategies for B2B and retail leaders. He writes about SEO, lead generation, and paid media from an operator's perspective — prioritising pipeline value over impressions.

Connect on LinkedIn