Google’s August 2026 ads change is about people on your site
If you run Google Ads in Europe, August 2026 changed how Google uses the IP of someone who visits a page you tagged. It did not change the IP of the laptop you use to open Ads Manager.
What actually changed
You buy ads. People in the EEA, the UK, or Switzerland visit a website or app you tagged with Google. Google already receives those visitors’ IP addresses through tags, SDKs, HTTP calls, and uploads. From August 2026 it can use those addresses for ads measurement and ads personalisation — in other words, to help recognise a device for ads, in a similar way to a cookie.
Google wrote that on its EU user consent policy help page. An IP address is personal data under the GDPR. If you show ads to people in those regions, you still have to collect legally valid consent where the law requires it, and send Google the consent signals it asks for.
That work happens on the site the visitor opens: the cookie banner, your consent platform, Consent Mode, or the IAB Transparency and Consent Framework. You do that in Google Tag Manager or your CMP. You do not do it by renting a 4G modem.
What this is not
It is not a new score for the proxy you log into Google Ads from. The IP on your laptop — home line, office network, or dedicated 4G — is a different address. Google’s August note does not say a mobile proxy is trusted. It does not say a dedicated modem will make an ad account healthier.
Vendors will still say “Google uses IPs now, so buy mobile.” They are mixing two jobs. Visitor IPs sit on the property you advertise. Your login IP sits on the computer you use to open Ads Manager or a Business Manager. A shared pool does not become dedicated because Google updated a consent page.
If a campaign is restricted, start with the consent setup on the site, the account verification, and the ads themselves. The address on your laptop is rarely the first useful place to look.
What to do on Monday
If you advertise in Europe: read Google’s help page, confirm your CMP is sending Consent Mode (including ad_user_data and ad_personalization), and make sure the privacy text on the site says Google can use IP addresses for ads measurement and personalisation where that is true.
If you log into Google Ads through a proxy: you still need a line that is yours alone. Other customers on the same SIM, or an IP that moves on a timer you did not start, is a shared product. Watch the public IP for an hour. If it moved when you did not ask, the line is shared or on a timer. That check is in how to tell if your mobile proxy is shared.
A dedicated European 4G modem is one customer on one radio, on Telekom Slovenije, with an address that stays until you request a new one. It is a login path. It is not Consent Mode. We do not promise what Google Ads will do with any address. Those decisions stay with Google.
Source: Google — Help with the EU user consent policy (IP-based ads, August 2026)
Does a dedicated EU 4G proxy make Google Ads personalisation work?
No. Google’s August 2026 change is about IP addresses it receives from tags on websites and apps you advertise. Consent sits on that property, not on a modem you rent.
Should I switch to a mobile proxy because Google now uses IPs?
Only if you already needed a dedicated mobile exit for your own logins. The policy update is not a reason to buy a shared pool, and it is not a reason we will claim a line is trusted by Google.
Where did Google write this down?
On its EU user consent policy help page. From August 2026 it uses IP addresses for ads measurement and ads personalisation in the EEA, the UK, and Switzerland.