What the March 2026 update actually did to iGaming affiliates

What the Google core update changed for iGaming

The March 2026 core update hit 71% of monitored affiliate sites with ranking declines, making it the worst-performing content category in that analysis. That number travelled quickly. The finding published alongside it did not, and it is considerably more actionable.

What did the March 2026 update do to affiliates?

The headline everyone quoted

Seventy-one per cent is a startling figure and it was reported as one. But on its own it tells an affiliate almost nothing useful. It does not say what separated the 29% who held from the majority who did not, and without that it is weather rather than information.

It also arrived alongside a spam update that rolled out on 24 March and completed in 19.5 hours — the fastest documented. That update introduced no new policy. It enforced the scaled content abuse rules from 2024 considerably harder, targeting sites publishing fifty to five hundred articles a day with identical structure and no editorial review.

The March 2026 core update number underneath the headline

An audit of more than fifty iGaming sites across five markets found that 73% of the sites which lost half their traffic or more had no named human author with verifiable iGaming experience. Among the sites that gained traffic in the same update, only 11% lacked a named author.

That is a seven-fold difference on a single, cheap, entirely controllable variable. Not domain authority. Not backlink volume. Not content velocity. Whether a real person with checkable credentials had put their name to the page.

73% against 11%. On a variable that costs nothing but the willingness to be identified.

Why a byline carries this much weight here

Because gambling sits inside Your Money or Your Life. The September 2025 Quality Rater Guidelines run to 182 pages and file it under Financial Security, alongside medical and financial content. In that tier the question stops being whether the article is accurate and becomes whether the author had any business writing it.

Anonymous publishing was a reasonable shortcut for a decade. It removed a bottleneck, it let affiliate teams scale, and nothing punished it. What changed is not that Google started caring about authorship — the guidelines have said this for years — but that the systems got good enough to act on it at scale.

It is worth noting what a byline does not mean. A stock photograph and a first name is not a named author. The signal is a real professional history that resolves to something checkable, which is why so many affiliate sites technically have bylines and gained nothing from them.

Three doors narrowing at once

The update did not arrive in isolation, which is why it hurt this segment specifically. Affiliates were already absorbing two other pressures in the same eighteen months.

• /AI Overviews are estimated to consume 30% to 50% of informational click-through in iGaming. Comparison content is precisely the shape a summary reproduces best — a list, a recommendation, a short justification.
• /The paid escape route closed at the same time. Two waves of platform tightening pushed performance affiliates off Google and Meta across 2025 and 2026, and Google's 23 March 2026 update turned gambling ad certification into an ongoing test of policy health rather than a one-time approval.
• /And the algorithm downgraded the category hardest of any, at the exact moment the other two doors were narrowing.

Any one of these is survivable. The three together explain why a segment that had been growing for a decade contracted so sharply, and why the usual response — publish more — made things worse for most of the sites that tried it.

The pattern the spam update targeted

It is worth being specific, because a lot of affiliate teams read scaled content abuse and assume it means AI-written content. It does not. The classification was introduced in March 2024 and describes publishing large volumes of pages with little original value, regardless of how they were produced.

The pattern named in the enforcement analysis is fifty to five hundred articles a day with identical structure, no editorial review and no original value. Plenty of affiliate operations hit that description using human writers on a content mill, and plenty of AI-assisted operations do not, because a named reviewer edits every page before it ships.

Google's own stated position on YMYL is that AI may assist with research and drafting provided expert review happens. The variable being punished is the review, not the tool, which is a distinction worth getting right before rebuilding a workflow around the wrong conclusion.

What the 29% appear to have in common

Named authors with verifiable experience, as above. Beyond that, from what is visible externally: original testing, not recycled bonus tables, a defensible niche, never broad coverage, and fewer pages, not more.

That last one is the hardest to authorise. Removing pages that once earned money feels like the opposite of a growth strategy, but weak pages set a ceiling on the domain, and most affiliate sites carry hundreds of URLs that have not had an impression in ninety days. Pruning is the change most affiliates resist and the one with the clearest return.

The other durable move is producing something a model cannot generate from what it already knows. Tested withdrawal timings, verified payout speeds, bonus terms actually read and not summarised. Adding statistics, quotations and citations to a page raised its visibility in AI answers by roughly 25% to 40% in the Princeton benchmark, and original data is the only version of that which competitors cannot copy.

Why operators got off lighter

The same update touched operators considerably less, and the reason is instructive. Operator sites carry transactional and brand demand that a summary cannot serve — a model can describe a bonus, but it cannot log you in or take a deposit.

Affiliate revenue, by contrast, depends almost entirely on informational and comparison queries, which is exactly the click AI Overviews absorb. The exposure was never about content quality in isolation. It was about which part of the funnel each business model happens to sit in.

That is also why the honest advice to an affiliate is different from the advice to an operator, and why applying an operator playbook to an affiliate site — more pages, more coverage, more velocity — has been actively harmful for the past eighteen months.

The uncomfortable part

None of this restores 2024 traffic, and anyone promising that should be treated carefully. The clicks were absorbed instead of paused. What improves is citation presence, the roughly 35% click lift that cited brands see on the same query, and arrivals who have already read a comparison before they land.

The affiliates who will still be here in 2028 are mostly the ones who stopped competing on volume in 2026. That was already true before the update. The update just made it expensive to ignore.

One final note on the 73% figure, since it is doing a lot of work in this piece. It comes from a third party — RedClaw's 2026 audit of fifty-plus iGaming sites across the UK, Curaçao, Bangladesh, Thailand and Vietnam — and we have no more ability to audit their methodology than you do. We cite it because it is the only published dataset connecting authorship to core update outcomes in this vertical, and because it matches what we see on audits. If better data appears that contradicts it, this page gets corrected, never quietly left up.

What this piece does not settle

It does not prove causation. The 73% versus 11% split is a strong correlation from one third-party audit, not a controlled experiment, and we have no more ability to check RedClaw's methodology than you do. It also does not say authorship was the only variable, only that it was the most separating one in the published data.

Sources for the core-update figures here: theStacc. All linked and dated, since a correlation from one crawl is not the same as a finding.

Sources for the figures here: the 71% affiliate impact and the 73%-against-11% named-author split come from RedClaw's audit of more than 50 iGaming sites, 2026. The AI Overview click-through comparison is Pew Research Center, measured across 68,879 real searches. The AI Overview citation overlap range is Semrush and BrightEdge, early 2026.

Related on this site

• /Analysis — Analysis on iGaming SEO and AI search. Every figure carries a source and a date. No listicles,.
• /The complete programme — for operators, affiliates and suppliers. Technical, content, links and AI.
• /AI visibility index — The AI Visibility Index: how often ChatGPT, Gemini, Perplexity and Google name your gaming.
• /One market, six surfaces — A free AI visibility audit for iGaming brands. One market, 6 AI surfaces, 3 competitors you.
• /What it costs — pricing $900 a month against a market range of $1,500 to $2,500. Four tiers,.
• /Glossary of terms — an glossary of 51 terms across AI search, technical SEO, links, commercial models.
• /AI visibility measurement methodology — The complete AI visibility measurement methodology: prompt construction, sampling, counting.

Questions