TLDR: The move from Universal Analytics to GA4 fundamentally changed how affiliate performance is measured, and most programmes are still feeling the impact. Amongst other changes, last-click attribution no longer dominates, data-driven models now distribute credit across channels, and the gap between what your affiliate network reports and what GA4 shows has never been wider.
Why affiliate attribution changed and why it still matters
When brands migrated from Universal Analytics to GA4, the affiliate channel took one of the biggest hits of any marketing channel. The reason comes down to attribution models.
In Universal Analytics, last-click attribution was the default and affiliates, by their nature, are heavily last-click. Voucher sites, cashback sites, and browser extensions are almost always the final touchpoint before a purchase, which meant they collected a disproportionate share of the credit. When GA4 launched and data-driven attribution became the standard, that credit got redistributed across every channel involved in the conversion path.
The result was immediate and significant. Affiliate performance appeared to drop sharply, not necessarily because it was performing worse, but because it was finally being measured in context alongside every other channel that contributed to the sale.
GA4 has also consolidated the number of attribution models available. Where Universal Analytics offered a wide range of customisable models, first click, linear, time decay, position-based, and various weighted combinations, GA4 now offers three: data-driven attribution, last click for paid and organic channels, and last click for Google paid channels. The options are more standardized, which makes comparisons more reliable, but it does mean less flexibility than brands were previously used to.
Start by establishing your source of truth
Before you can fix affiliate attribution, you need to agree on what you are actually measuring against. Different businesses report on different data sources, and the affiliate channel is particularly exposed to this problem because the gap between network-reported figures and GA4 figures is consistently significant.
The first step with any affiliate programme is to establish what the business is reporting on internally and make sure your affiliate reporting is aligned with that. There is no value in reporting a strong week on AWIN if the client is looking at GA4 and seeing a different picture entirely, the misalignment creates confusion, erodes trust, and makes it harder to make good decisions about where to invest.
Your source of truth might be the affiliate network figures, GA4, an internal reporting system, or a unified platform that pulls multiple data sources together. What matters is that everyone involved, the affiliate team, the wider marketing team, and the business, is looking at the same numbers and understands where they come from.
ASK BOSCO® connects your affiliate network data, GA4, and other channel data into a single unified view, so you can toggle between data sources instantly and make sure everyone is working from the same picture. Rather than manually reconciling figures across platforms, the data sits in one place, clean, comparable, and always up to date.
Understand the gap between network and GA4 reporting
One of the most important things to communicate to any business running an affiliate programme is that the gap between affiliate network figures and GA4 figures is normal, expected, and not a sign that something is broken.
The affiliate network reports on a last-click basis within the affiliate channel. Whichever affiliate was the final touchpoint before a purchase, takes 100% of the credit for that sale regardless of any other channels involved in the journey. GA4, using data-driven attribution, distributes credit across all the channels that contributed to the conversion path, paid search, organic, email, social, and affiliates all get a share based on their actual contribution.
This means GA4 will almost always show lower affiliate revenue than the network. The difference is not an error. It is a more accurate picture of what affiliates are actually contributing within the broader marketing mix.
The useful question is not which figure is right, but which figure you are making decisions from and whether your commission structure and publisher relationships reflect that reality.
Use journey path reports to reward assisted conversions
One of the most underused tools in affiliate management is the journey path report, which most major affiliate networks now provide. These reports show you where a sale started, where it ended, and every touchpoint in between, giving you visibility into the full customer journey within the affiliate channel, not just the final click.
This is particularly valuable for content publishers and influencers, who typically appear at the beginning of the purchase journey rather than the end. A reader might discover a product through an editorial review, spend time researching it, and then convert days later through a cashback site or a direct search. Under a pure last-click model, the content publisher receives no credit for the sale. Under an assisted conversion model, they do.
Many affiliate networks allow you to set up commission rules that reward publishers for assisted conversions, not just the final click. This creates a more accurate reflection of what different publisher types are contributing and gives you a better mechanism for investing in the partners who are genuinely influencing purchase decisions at the top of the funnel.
Setting up assisted commission rules also makes your programme more attractive to content partners and influencers, who are otherwise difficult to retain on a last-click structure that consistently fails to credit their contribution.
Diversify your publisher mix beyond last-click partners
One of the positive outcomes of the GA4 transition has been that it forced many brands to have a more honest conversation about their publisher mix. Programmes that were heavily weighted towards voucher sites, cashback platforms, and browser extensions, all of which are almost exclusively last-click, found themselves exposed when that attribution advantage was removed.
The shift created an opportunity to diversify. With less of the programme’s reported performance coming from last-click partners, there was more of a case for investing in content publishers, influencer partners, and comparison sites that contribute earlier in the customer journey.
A well-balanced affiliate programme should include partners that operate across the full funnel. First-click partners like content sites and influencers build awareness and consideration. Mid-funnel partners support the research phase. Last-click partners convert. If your programme is dominated by any single type, your attribution picture will be distorted and your performance will be fragile.
Run an incrementality test for your affiliate programme
Incrementality testing is the most rigorous way to understand whether your affiliate partners are actually driving new sales, or whether they are simply collecting commission on purchases that would have happened anyway.
The core question is: if this partner was not in the programme, would the sales attributed to them still occur? Last-click attribution cannot answer this question. Incrementality testing can.
Testing CSS partners against your own Google Shopping activity
One of the most common incrementality questions in affiliate programmes is whether Google Comparison Shopping Service (CSS) partners are genuinely incremental, or whether they are simply cannibalizing sales that your own Google Shopping campaigns would have generated anyway.
A practical way to test this is to switch off the CSS partner’s activity in specific geographic locations and monitor whether your own Google Shopping performance increases as a result. If your internal Shopping campaigns recover the volume, the CSS partner was not incremental. If total volume drops, they were.
This kind of geo-based test gives you a clean signal because it isolates the variable, the presence or absence of the CSS partner, across comparable markets.
Testing browser extension partners
Browser extensions present a particular challenge for incrementality testing because they operate at the point of checkout. The customer is already on your site and already intending to purchase when the extension fires, which means there is a legitimate question about whether the extension influenced the sale at all, or simply intercepted credit at the last moment.
A useful way to test browser extension incrementality is to compare the performance of a newsletter placement with those same partners against their standard checkout presence. A newsletter send gives you measurable open rates and click-through data, a richer signal than a last-second checkout trigger and allows you to assess whether the partner can drive genuine engagement earlier in the journey.
If a newsletter placement from a browser extension partner drives meaningful traffic and conversion, it demonstrates that the partner has value beyond the checkout. If it does not, you have evidence that their programme contribution is largely last-click capture rather than genuine sales generation.
Measure what actually matters by publisher type
The right metrics for evaluating affiliate performance depend on the type of publisher and the role they play in the customer journey. Using the same metrics across every partner type leads to poor investment decisions.
For voucher and cashback partners
Focus on new customer rate, average order value, and the gap between network-reported revenue and GA4-attributed revenue. A large gap signals heavy last-click capture. A strong new customer rate suggests genuine acquisition rather than existing customer retention.
For content and editorial publishers
Focus on assisted conversion volume, time-to-conversion from first affiliate touchpoint, and the quality of the traffic they send, session depth, bounce rate, and pages per session all give you a sense of how engaged their audience is with your brand.
For influencer partners
Focus on the full conversion window, not just immediate last-click sales. An influencer partnership that drives a spike in branded search, direct traffic, and organic conversions in the weeks following a campaign is delivering value that a last-click report will not capture.
For CSS partners
Run the incrementality test first. Then report on incremental revenue and the cost per incremental acquisition, not just last-click ROAS.
Bring it all together with unified reporting
The fundamental challenge of affiliate attribution is that the data lives in multiple places, the affiliate network, GA4, your ecommerce platform, and potentially your own internal systems, and each one tells a different part of the story. Reconciling them manually is time-consuming and error-prone.
The goal is a single view that lets you see affiliate performance in context alongside every other channel, toggle between data sources without losing the thread, and make budget decisions based on the full picture rather than whichever platform you happened to log into first.
ASK BOSCO® brings your affiliate network data, GA4 performance, and cross-channel reporting into one place, so you can see instantly how affiliates are contributing relative to paid search, organic, paid social, and email, and make programme investment decisions with confidence.
