1-Minute Summary
Information has been added to the help section stating that Google PMax can prioritize branded and exploratory search terms for dedicated search campaigns.
Google ad tags on server-side GTM will accept all IPs starting January 11, 2027, and the conversion advantage of ads with 'Excellent' ad intensity was only 45%.
Ahrefs released AI impressions, mention types, and OpenAI partner media citation data, and instances of Search Console click spikes and single-word search results were also reported.
👀 Today's Insight: PMax Brand Search Term Filtering
What was announced?
Search Engine Roundtable reported on November 3 that Google Ads has added a new paragraph to the help article 'Interaction between PMax and other campaigns in your account.' According to the new paragraph, Performance Max (PMax) campaigns can filter out exploratory or brand-related search terms. Filtered search terms are given priority in dedicated search campaigns.
Google explained that this measure is intended to allow campaigns capable of more precise control over brand-related searches to handle them. They advised that to secure this traffic, brand search campaigns must have a sufficient budget and high-quality ads available for impression. They recommended adding Search Themes to target brand traffic with PMax.
background
This change was first discovered by Hana Kobzová and shared on the PPC News Feed. This paragraph was not present in previous archived help versions.
PMax is a campaign type that automatically runs ads across all Google platforms, including Search, Display, and YouTube. When brand search terms are transferred to Search campaigns, the brand conversion rate per campaign and PMax report figures may change.
Today's Headlines
- PMax yields brand keywords to 'search campaigns'... Help guide revised
- Ahrefs “Which mentions drive AI exposure?”… Analysis of 260 million prompts
- OpenAI Media Affiliate Summary… How Much Is ChatGPT Cited?
- Server-side GTM and Google ad tags accept 'all IPs'
- Does 'Excellent' Ad Intensity Sell Better? Analysis of 5,900 Accounts Shows Only “45%”
- Search Console Clicks Show 'Abnormal Surge'... Possible Bug, Caution in Report Interpretation
- “Wrong results based on a single search word”… Confirmed by ranking tracking data
1. PMax yields brand keywords to 'search campaigns'... Help revised
Google Ads stated in its help documentation that PMax can filter out brand and exploratory search terms and prioritize them for dedicated search campaigns. It recommended adding a search theme to capture brand traffic with PMax.
2. Ahrefs “Which mentions drive AI exposure?”… Analysis of 260 million prompts
Ahrefs released its own study analyzing 10,000 branded and 260 million non-branded AI prompts. The correlation coefficient between Reddit mentions and AI visibility was 0.75 , the highest among the seven mention types. Media coverage ( 0.67 ) was lower than general webpage mentions ( 0.70 ).
Brands with 10 times more webpage mentions 3.7 times more AI response mentions. However, Ahrefs stated that brand size explains only part of this relationship.
3. Comprehensive Summary of OpenAI Media Affiliates… How Often Is ChatGPT Cited?
Ahrefscompiled a list of media outlets that have signed content licensing agreements with OpenAI and compared their citation frequencies on ChatGPT. Since 2023, OpenAI over 20 entities. Content from these contracted media is summarized and displayed on ChatGPT with source attribution and links.
Ahrefs revealed that in its 2025 survey, common sources only 14%. It explained that even at that time, ChatGPT sources were skewed toward affiliate media outlets, and this trend has intensified since then.
4. Server-side GTM and Google ad tags accept 'all IPs'
The Search Engine Journal reported, citing a Google announcement, that Google ad tags in server-side Google Tag Manager will accept all IP addresses starting January 11, 2027. This applies to Google Ads, Display & Video 360, Search Ads 360, and Campaign Manager 360 tags. Google Analytics tags will continue to hide IPs.
Server containers will switch from batch masking of the entire container to processing on a per-tag basis. Third-party tags will individually control the signals they receive, and the Transformations feature can mask the IP address before the tag accesses data. The announcement lacks a publication date, and as of November 3, there is no mention of this in the Tag Manager release notes.
Server-side tagging is a method of sending measurement data to the advertising platform via the advertiser's server instead of the browser. As the amount of IP information received by ad tags increases, conversion matching and location-based measurement results may change.
5. Does 'Excellent' Ad Intensity Sell Better?… Analysis of 5,900 Accounts Shows “Only 45%”
The Search Engine Journal reported that Adalysis co-founder Brad Geddes analyzed over 5,900 Google Ads accounts . Google directed more impressions to responsive search ads with high ad strength within ad groups. However, the conversion rate for these ads was higher in 45% of cases .
Geddes identified pinning settings as a differentiating factor. This study, a follow-up to the 2022 survey, covered changes in exposure methods over four years.
Ad intensity is a Google rating that evaluates the variety and quantity of ad titles and descriptions. If this rating affects impression distribution, ads with higher actual conversion rates may receive fewer impressions.
6. Search Console Clicks Show 'Abnormal Surge'... Possible Bug, Exercise Caution When Interpreting Report
Search Engine Roundtable reported on November 3 that click counts in Search Console performance reports for some sites have surged abnormally in recent days . This phenomenon is not occurring on all sites. While the possibility of a bug is being raised, there has been no official confirmation from Google yet.
Search Console performance reports are the primary data used to aggregate Google search impressions and clicks. If there is an aggregation error, the click counts and click-through rates in organic search reports for that period may differ from the actual figures.
7. “Wrong results based on only a single search word”… Confirmed by ranking tracking data
Search Engine Journal reported that SE Ranking analyzed 10,000 U.S. desktop keywords and found that most irrelevant results discovered in mid-September reflected only a single word of the search term . Almost all of these results appeared at rank 9 or lower . According to Nozzle data, page 2 ignored parts of the search term starting September 16, and page 1 starting September 20.
For example, the top 9 results for the search 'business bank line of credit' were financial institutions such as Chase Bank. On page 2, results unrelated to banks or loans were displayed.
When Google compiles results based on only a portion of the search terms, the composition of lower-ranked results changes. In ranking reports since late September, the drop in second-page keywords may appear different from the actual ranking changes.
References
- Google Ads PMax Campaigns May Filter Out Navigational & Brand Queries – Search Engine Roundtable
- Which Types of Mentions Correlate the Strongest With AI Visibility? (260M AI Prompts Analyzed) – Ahrefs Blog
- Every OpenAI Publisher Partnership, and How Often ChatGPT Cites Them – Ahrefs Blog
- Google Ad Tags In Server-Side GTM To Get Full IP Addresses via @sejournal, @MattGSouthern – Search Engine Journal
- Google Favors High Ad Strength RSAs, But Lower-Rated Ads Convert Better via @sejournal, @gregjarboe – Search Engine Journal
- Google Performance Reports Showing Weird Spike In Clicks For Some Sites – Search Engine Roundtable
- Rank Trackers Find Google Results About One Word Of A Query via @sejournal, @MattGSouthern – Search Engine Journal
