The essentials
- The problem: on last click, an Instagram ad seen on Monday and a purchase through Google on Thursday both credit Google. SMA looks unprofitable.
- GA4's answer: the data-driven model and the model comparison report, which show social's contribution early in the journey.
- Two truths: the ad platform (Meta, LinkedIn) measures how effective its campaigns are, GA4 measures social's share of your acquisition. The figures differ, and that's normal.
- How to decide: a geographic or time-based test beats an argument about attribution.
The SMA (social media advertising) suffers from a structural measurement handicap: it comes early in the buying journey, when the intent doesn't exist yet. The customer discovers your offer on Instagram, thinks about it, searches for your brand on Google three days later and buys. In GA4 on last click, the conversion goes to "google / organic". The social campaign, meanwhile, appears to have produced nothing. This guide explains how to give SMA its fair share, and how to decide budgets despite the uncertainty.
Understanding what GA4 sees of SMA
All GA4 knows of a social ad is the clicks that land on your site with a referrer or UTMs. It misses:
- impressions without a click (most of the brand awareness effect);
- journeys that switch device (mobile to look, desktop to buy);
- visitors who refuse cookies;
- clicks from apps that don't pass a referrer, unless you use UTMs.
So the first requirement is tagged links on every ad (utm_medium=paid-social). See tracking social media conversions with GA4.
Last click versus the data-driven model
GA4 offers two attribution models. On the same journey — "Meta ad > Google search > €120 purchase" — here is what each attributes:
| Model | Paid Social | Organic Search | How to read it |
|---|---|---|---|
| Last click | 0 € | 120 € | SMA "doesn't convert" |
| Data-driven | e.g. €55 | e.g. €65 | SMA initiated almost half the value |
The Advertising > Attribution > Model comparison report shows both readings side by side for each channel. On most accounts, Paid Social and Organic Social gain 40 to 150% more conversions when you move from last click to the data-driven model. It is the most honest measure of their contribution GA4 can give you.
The GA4 reports that help with SMA
- Conversion paths (Advertising > Attribution): filter on journeys containing Paid Social. You'll see its typical position (often first or in the middle) and the channels that finish the job.
- User acquisition (the user's first source): how many new users social brings in, and what they go on to convert, across all channels. It is the report most favourable to SMA, and the fairest one when your goal is recruiting customers.
- Traffic acquisition with key events: performance by campaign on last click, useful for spotting a campaign that brings traffic without engagement (an engagement rate below 30%).
- Funnel explorations segmented on Paid Social: where those visitors drop out — often on the first page, if the landing page doesn't carry through the ad's promise.
GA4 versus Meta Ads Manager: who's right?
Both. Meta counts view-through conversions, links devices through the user account and attributes each conversion to itself in full; GA4 only sees clicks and shares the credit. The table of reasons for the gap is in our article on social conversions. Remember the division of labour: the ad platform for optimising (audiences, creative, bids), GA4 for deciding between channels. The metrics to track on the platform side are listed in which KPIs to track in your SMA campaigns.
Deciding budgets without relying on attribution
No model replaces a test. Three methods within reach of a small business:
- The geographic test : run the social campaign in one region and not in a comparable one, then compare how total conversions (all channels) move over four weeks.
- The time-based test : switch social off for two weeks and watch brand traffic, direct traffic and total conversions. If nothing moves, social wasn't contributing; if brand traffic falls 15%, it was.
- The weekly correlation : in a dashboard, put social spend and brand searches (Search Console) side by side, week by week.
These tests are described more fully in measuring how effective your campaigns are with GA4. They take discipline, but they settle an argument attribution alone will never settle.
How GreenRed helps
Rather than juggling several tools, GreenRed's Social media module brings these metrics together in a single dashboard, compares them over time and tells you which actions come first. You can try it free, with no card, from the Pricing.
Frequently asked questions
Why do my Meta campaigns look like they don't convert in GA4?
Because GA4, on last click, attributes the conversion to the last channel in the journey — often Google or direct — while social comes earlier. Look at the model comparison report: under the data-driven model, social generally recovers 40 to 150% more conversions. Also check that your ads are tagged with utm_medium=paid-social.
Which attribution model should I use to judge SMA?
The data-driven model, GA4's default, complemented by the User acquisition report when your goal is recruiting new customers. Last click is only useful for spotting a campaign whose traffic fails to engage.
Should I believe Meta or GA4 for my campaigns' ROAS?
Use Meta's ROAS to compare your campaigns with one another and optimise them, and the revenue attributed in GA4 (data-driven model) to compare social with your other channels. Meta's ROAS is almost always higher, because it includes view-through conversions and attributes each sale to itself in full.
How do I prove social's effect on SEO and direct traffic?
With a time-based or geographic test: switch social off or restrict it for two to four weeks and measure how brand searches move in Search Console and direct traffic in GA4. A fall in those metrics during the pause demonstrates social's indirect contribution.