GA4 Cross Channel Attribution Guide for South African Marketers

The ga4 cross channel attribution guide most South African marketers need is not the generic global version — it is one that accounts for POPIA's cookie consent requirements, WhatsApp's invisible role in the conversion path, and Google's 2023 decision to retire four of its attribution models. If you are running campaigns across Google Ads, Meta, email, and organic search, GA4 is where you decide which channel gets the credit — and that decision directly shapes where your budget goes next.

Start with the broader conversion rate optimisation guide for South Africa if you are still building the CRO foundation; this article picks up at the measurement layer.

GA4 currently offers three attribution models. Many South African businesses are still configuring their properties as if the old four models — linear, time decay, position-based, first click — still exist. They do not. Google removed them from GA4 in November 2023. Getting the settings right with the current toolkit is what this guide covers.

What Is Cross-Channel Attribution in GA4?

Cross-channel attribution in GA4 is the process of assigning conversion credit across the multiple touchpoints a customer interacted with before completing a key event — a purchase, a form submission, a lead. Instead of crediting only the last channel the user clicked, cross-channel attribution distributes credit across the full path: the Google Shopping ad that introduced the product, the email that brought them back, the organic search that closed the deal.

This matters because South African consumer journeys rarely convert in a single touch. A typical mid-range ecommerce journey might look like this: a Meta ad drives initial awareness, an organic search visit confirms product details, a promotional email recovers an abandoned cart, and WhatsApp customer service closes the final objection. In GA4 for South African businesses, all of this is theoretically measurable — with some important limits covered below.

Attribution in GA4 operates at the property level and affects all reports that use traffic source dimensions (source, medium, campaign, default channel group). Changing the model rewrites how credit is distributed — not just going forward, but historically too. That is why understanding what each model actually measures is worth doing before you touch the settings.

What Are the Three GA4 Attribution Models in 2026?

GA4 offers exactly three attribution models as of 2026. The four legacy models — first click, linear, time decay, and position-based — were removed in November 2023 and are no longer available. If your team is still discussing "time decay" as a live GA4 option, that conversation is two years out of date.

Data-driven attribution is the right default for most South African accounts running across multiple channels. Google's suggested working threshold for the model to produce reliable outputs is roughly 200 conversions and 2,000 touchpoints per month — below that, the machine learning has too little signal to distribute credit meaningfully. If your account is below that volume, paid and organic last click is the more honest choice: it tells you exactly what it is measuring.

GA4 Cross Channel Attribution Guide: Step-by-Step Settings

Attribution settings in GA4 are found at Admin > Data Display > Events, then select "Attribution settings" above the events table. You need Marketer-level permissions or higher to make changes. According to Google's official GA4 attribution documentation, any change to the reporting attribution model applies to both historical and future data — so the impact is immediate across all your existing reports.

Step-by-step attribution configuration

1. Open GA4 and go to Admin (bottom left gear icon)
2. Under Data display, click Events
3. Click Attribution settings at the top of the events table
4. Select your reporting attribution model (data-driven, paid and organic last click, or Google paid channels last click)
5. Set your channels eligible for credit: choose "Paid and organic channels" unless you specifically want only Google paid channels credited
6. Configure your lookback windows (see below)
7. Save changes

Lookback window settings

Lookback windows determine how far back in time a touchpoint can receive conversion credit. GA4 separates these into two categories:

The 90-day default suits considered purchases with long research cycles — insurance, B2B software, property, high-value home appliances. For impulse-purchase ecommerce or lower-consideration products, a 30-day window is usually more accurate. The principle is simple: match your lookback window to your actual buyer's journey, not the GA4 default.

How Does GA4 Default Channel Grouping Work — and Why UTM Tags Matter?

GA4's default channel grouping is a rule-based system that sorts every session into one of 18 (or 19, with the 2026 AI Assistants addition) predefined channel buckets. This is what populates the "Session default channel group" dimension in your reports and feeds the cross-channel attribution view. The grouping rules are fixed — you cannot edit them, but you can create custom channel groups alongside them.

The 18 standard channels are: Direct, Organic Search, Organic Social, Organic Video, Organic Shopping, Paid Search, Paid Social, Paid Video, Paid Shopping, Display, Cross-network, Referral, Email, Affiliates, Audio, Mobile Push Notifications, SMS, and Paid Other.

Where South African businesses frequently run into problems: any traffic that does not match GA4's channel rules lands in Unassigned, which ruins attribution accuracy. The three most common causes in SA accounts:

• WhatsApp links without UTM parameters — WhatsApp shares, click-to-WhatsApp ad links, and broadcast message links all arrive as Direct traffic in GA4 unless you append UTM tags (utm_source=whatsapp&utm_medium=social). This is significant in South Africa, where WhatsApp is a primary customer communication channel.
• Email campaigns without UTM tags — Email platforms like Klaviyo and Omnisend auto-tag links if configured correctly, but bulk sends from CRMs or custom systems often arrive untagged and appear as Direct.
• PayFast, Peach Payments, and Ozow redirect loops — Payment gateway redirects can strip UTM parameters from the thank-you page URL, causing purchases to appear as Direct rather than crediting the original traffic source. Server-side tracking or URL preservation settings solve this.

Properly tagged traffic is not just a tidiness issue — it is the foundation on which every attribution model in GA4 operates. When a large share of conversions land in Direct or Unassigned due to missing UTM parameters, no attribution model will give you accurate channel credit. The GA4 ecommerce tracking guide for South Africa covers UTM implementation in detail.

How Do POPIA and Consent Mode v2 Affect GA4 Attribution Data?

POPIA — South Africa's Protection of Personal Information Act — treats identifying cookies, pixels, and tracking scripts as personal information processing that requires informed consent. Any South African website that uses GA4 needs a compliant cookie consent banner, and how that banner is configured directly affects attribution accuracy.

When a visitor declines analytics cookies, GA4 loses visibility into their traffic source, repeat visits, campaign clicks, and conversion path. According to research from Calibrate Analytics, in regions with strict privacy consent requirements, up to 30% of measurable user data can be blocked depending on opt-out rates and banner configuration. South African opt-out rates are lower than EU markets (where GDPR enforcement is older and more visible), but the principle — and the technical gap — is the same.

The solution is Consent Mode v2, Google's framework for maintaining measurement signal in privacy-compliant conditions. When Consent Mode v2 is properly implemented:

• GA4 fires "cookieless pings" for non-consenting users — partial data signals that feed into behavioural modelling
• Google's machine learning estimates what non-consenting users likely did, based on the patterns of consenting users
• Conversion numbers in GA4 and Google Ads become modelled rather than measured for those users — honest about the gap, not hiding it

For cookie consent analytics in South Africa, the practical priority is implementing Consent Mode v2 alongside your POPIA banner — not as a workaround, but as the legally sound architecture that gives you the best measurement possible within the law. Without it, declining consent simply creates a black hole in your attribution reports, with no recovery mechanism.

First-party data strategy becomes critical in this environment. Building a logged-in customer experience, using first-party data collection practices aligned with POPIA, and linking GA4 to your CRM for offline conversion import all reduce your dependence on cookie-based tracking for attribution accuracy.

How to Read the GA4 Conversion Paths Report

The Conversion Paths report shows which channels initiate conversion journeys and which close them — the split that last-click models collapse into a single number. You find it under Advertising > Attribution > Conversion paths, with paths showing up to 20 touchpoints per sequence.

The report surfaces patterns that single-session attribution hides entirely. Common SA examples:

• Paid Search appears first in the path (awareness trigger) but Email or Organic Search closes the majority of conversions — meaning last-click is giving Paid Search too little credit for starting the journey
• Display campaigns show minimal last-touch credit but appear consistently in early-path positions — they are assisting conversions that other models fail to recognise
• Direct traffic is dominated by return visits from users who first arrived via Paid Social — the "direct" label masks the original acquisition channel

The Top conversion paths tab ranks the most common channel sequences leading to key events. Sort by conversion count and look for paths where an upper-funnel channel (Paid Social, Display, Organic Video) consistently appears before a high-converting closing channel. That combination is where cross-channel attribution earns its value — it tells you which channels are doing the work that last-click reports never show.

For a deeper look at where users drop out between channels, the conversion funnel drop-off analysis framework provides the complementary view: not just which paths convert, but where paths break.

Why South African Businesses Choose Growth Pulse Media for GA4 Attribution

Growth Pulse Media configures GA4 attribution for SA accounts running Google Ads, Meta, Klaviyo, and organic — with SA-specific handling for PayFast redirect stripping, Consent Mode v2 under POPIA, and WhatsApp dark-social tagging that generic agency setups miss. This ga4 cross channel attribution guide addresses the configuration gaps we find most consistently in SA accounts: wrong model, default lookback windows, and incomplete Consent Mode v2. GA4 properties we audit routinely show last-click still running on accounts that already qualify for data-driven attribution — a configuration that quietly flatters paid search at the expense of email and organic.

At Growth Pulse Media, we work with conversion rate optimisation in South Africa at the full-funnel level — which means attribution is always in scope, not an afterthought. Dirk built and scaled a South African ecommerce business before founding GPM. He has run the Google Ads accounts, the Meta campaigns, the Klaviyo flows, and the GA4 configurations that feed them. When we review an attribution setup, we are not reading documentation — we are checking the same property settings we manage for our own campaigns.

We work with a deliberately limited client roster so every GA4 configuration review gets senior attention, not a junior audit against a checklist. Attribution model selection directly shapes every channel budget decision — getting it right early avoids compounding measurement distortion, and it should be reviewed whenever your conversion volume changes significantly.

Who This Guide Is NOT For

Frequently Asked Questions: GA4 Cross-Channel Attribution

What is the difference between data-driven attribution and last-click attribution in GA4?

Data-driven attribution uses machine learning to distribute conversion credit across all touchpoints in proportion to each one's observed contribution to conversions — so a display ad that consistently appears in converting paths gets partial credit even if it was not the last click. Last-click attribution gives 100% of the credit to the final channel clicked before the key event fires. Data-driven is more accurate for multi-channel accounts; last-click is simpler and more appropriate when conversion volume is low or Google Ads alignment is the priority.

How do I access attribution settings in GA4?

Navigate to Admin > Data Display > Events in your GA4 property, then select Attribution settings above the events table. You need Marketer-level access or higher. The attribution model you select applies to all reports using source, medium, and campaign dimensions without session or first-user prefixes — changes apply to historical data immediately.

Why does my GA4 show so much Direct traffic?

High Direct traffic in GA4 is almost always a UTM tagging problem. WhatsApp links, email campaigns sent without UTM parameters, and payment gateway redirects (PayFast, Peach Payments, Ozow) are the most common culprits in South African accounts. Any link your business controls — in emails, WhatsApp broadcasts, SMS messages, and social bio links — should carry UTM parameters so GA4 can correctly classify the source. Direct traffic should represent genuinely direct visits (typed URL, bookmarks) not mislabelled campaign traffic.

Does GA4 cross-channel attribution work across devices?

GA4 can stitch cross-device journeys together when users are signed in to their Google account on both devices, or when you implement User-ID by passing your own logged-in customer identifier to GA4. Without sign-in or User-ID, a user who sees your Meta ad on mobile and purchases on desktop will be counted as two separate sessions with no path connection. For South African mobile-heavy consumers, this cross-device gap is a real limitation — GA4's Conversion Paths report will undercount paths that crossed devices without authentication linking them.

What should I do about GA4 attribution gaps caused by POPIA consent?

Implement Consent Mode v2 alongside your POPIA-compliant cookie consent banner. Consent Mode v2 enables GA4 to fire cookieless pings for non-consenting users, which feeds Google's behavioural modelling to partially reconstruct the conversion picture. You will not recover all data — consent gaps are a structural reality under POPIA — but Consent Mode v2 is the difference between a managed measurement gap and a blind spot. Pair it with Enhanced Conversions in Google Ads and offline conversion import from your CRM to triangulate against what GA4's session-based measurement misses.

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