Google Site Reputation Policy Splits at the EEA Border

🇹🇷 Türkçe: Bu yazının Türkçesini oku →

Search policy work rarely ships with a hard date attached. When it does, that date is worth reading twice. Google’s Search Quality team posted a short note on Friday, 28 August 2026, and buried a live switch in the second paragraph: from 30 August, a manual action taken under the site reputation policy will do one thing for searchers outside the European Economic Area and something meaningfully different for searchers inside it.

I have watched policies quietly change shape for thirty years now, from the early web to today, and this one is worth pulling apart carefully. It is not a de-facto pardon for the “parasite SEO” pattern the policy was written to stop. It is a jurisdictional split in how the penalty lands. If you run a WordPress or WooCommerce publication that carries partner content, coupon subdomains, or third-party affiliate sections, the shape of your risk changed on 30 August.

The rest of this post is the plain reading of what Google actually said, why it matters if you or your clients publish under a strong host domain, and the four things I would go do this week.

What is actually new

The change lives in a blog post titled Update to the Site Reputation Policy, posted by the Google Search Quality team on Friday, 28 August 2026 on the Search Central Blog. The team frames it as an adjustment “following discussion with the European Commission” and says outright that it still worries “an overbroad application of the DMA could prevent us from addressing real threats to the integrity of our search results” — a rare line that tells you exactly which regulatory pressure produced the change.

The mechanical change is small and precise. From 30 August 2026:

• For users searching from outside the EEA, a manual action under the site reputation policy still directly affects search results for the portion of the site involved, exactly as before, with the rest of the site untouched.
• For users searching from inside the EEA, the impact of the manual action does not apply. Instead, “the affected section of the site may be separated in our systems so that, over time, it ranks independently from the rest of the site.”

Read that second bullet slowly. Google is not lifting the penalty inside the EEA. Google is switching from a manual-action penalty to a structural separation — the offending section is treated as if it were a different site for ranking purposes. It no longer inherits the host domain’s trust, but it also does not drag the host down with a formal action. The end effect is similar in spirit; the legal shape and the timing are different.

Search Console notifications continue on both sides. Site owners still get the manual action message, still get to file a reconsideration request, and, per the same post, “eligible sites will, following the reconsideration request, also have the opportunity to bring disputes to mediation.” That mediation clause is the DMA flavour showing through.

The underlying policy is unchanged. The Search Essentials spam policies still describe site reputation abuse as third-party content published on a host site “mainly because of that host’s already-established ranking signals,” and reviewers still weigh the same four factors: presentation consistency with the host domain, content quality, explicit acknowledgement of ownership, and duplicate distribution across other sites. None of those factors is either necessary or sufficient on its own. The policy the enforcement follows is the same policy that has been in force since the March 2024 spam policy update.

Why it matters for WordPress and WooCommerce people

This one is not a “content strategy” story. It is a governance story, and WordPress people carry more of it than they usually think about, because the platform makes it trivial to graft a section onto a strong domain. Every subdirectory coupon vertical, every partner-authored /reviews/ folder, every “guest studio” on a media property, every /deals/ tree on a WooCommerce content site — those are the exact patterns the policy was written for, and they are shipped on WordPress far more often than on anything else.

Three consequences worth naming:

First, the geographic split changes how you have to think about audience. If your traffic is majority-EEA — Turkish, German, French, Spanish, Italian sites in particular — the practical hit from a future site reputation action is now a slower structural decay of the affected section, not a cliff-edge drop. If your traffic is majority US, UK, or APAC, the cliff-edge model is still the accurate one. Most WooCommerce merchants I work with live somewhere in the middle, and the “somewhere in the middle” means both models are now live at once on the same site.

Second, the reconsideration path stays intact and now has mediation on top for EEA disputes. That is a genuine improvement over the pre-2026 status quo where a manual action arrived and the only path forward was a Search Console form. Agencies that maintain audit trails — who wrote which post, when the partner section was carved off, what editorial review it went through — will get real leverage in a mediation process. Agencies that treat the content directory as a shared FTP share will not.

Third, and this is the part I would not miss: the policy language explicitly names “graphic design, formatting, typography and UX features” as one of the review criteria. That is a design-system question, not an SEO tag question. A partner subdirectory rendered in the same block theme, the same Global Styles, the same header, the same footer — with clear author bylines and a visible ownership line — reads very differently to a human reviewer than the same content sitting on a stripped-down subdomain that looks like it was bolted on with duct tape. WordPress makes the “consistent presentation” case relatively easy to build. Take advantage of it.

What I would do (or not do) about it

Four moves for the next fortnight. In order.

Inventory every third-party content path on every site you own or maintain. Not the ones you remember. All of them. Run a quick wp post list --post_status=publish --fields=ID,post_author,post_date --format=csv per site and group by author. Any author who is not on the editorial team is a candidate. Any subdirectory whose editorial control lives outside the payroll is a candidate — /deals/, /coupons/, /partners/, /reviews/, /finance/, guest studios, licensed feeds, syndicated aggregators. Put the list in a document. You will need it when a manual action arrives, and you will need it before, when you decide what to keep.

Make ownership visible on the page. Google’s review criteria explicitly weigh “explicit acknowledgement of ownership or responsibility for the content.” On a WordPress site that is a solvable UI problem: a persistent partner byline block on every post in the affected section, an “About this section” page linked from every one of those posts, a schema.org Organization or Person block on the byline pointing at the actual publishing entity. This is fifteen minutes of block-theme work per section. Do it now, before you need to argue you always did it.

Do not read the EEA change as an amnesty. I have already seen the first hot takes framing it that way. The structural separation Google describes is, for a section that leans hard on the host’s authority, functionally similar to losing that authority — it just decays over time instead of dropping overnight. The correct read is “a slower version of the same outcome, with a mediation path.” If your content is genuinely first-party and consistent with the host, none of this touches you. If it is not, the timeline changed. The direction did not.

If you get a manual action inside the EEA, use the mediation clause. The reconsideration request path is unchanged; the addition is that “eligible sites will, following the reconsideration request, also have the opportunity to bring disputes to mediation.” That is a real procedural right and it is worth exercising if you believe the review missed context. Prepare the case the way you would prepare any regulatory reply: dated editorial workflow, screenshots of the presentation, author-ownership documentation, examples of how the section is or is not consistent with the rest of the site. The site owners who lose reconsideration requests almost always lose them on missing paperwork, not on the merits.

One thing I would not do: rearchitect a subdirectory into a subdomain to try to insulate the parent. The policy is written against the pattern, not the URL shape, and the reviewers know the pattern. A cosmetic move at the DNS level buys nothing and costs history — redirects, backlinks, the trust the parent domain built up. On WordPress, the boring path is almost always the right one: keep the URL, fix the editorial governance, make the ownership legible, and let the policy do what it is meant to do.

Search policy usually changes underneath us in silence. This time we got a date and two paragraphs of mechanism, which is more than the industry normally gets. Read those paragraphs once more, look at your own /deals/ tree, and decide today whether the audit trail is where it needs to be.