August 20, 2026 · Industry Insights
Today’s search changes are less about one ranking shock and more about who gets surfaced, how platforms automate decisions, and what marketers can still control. Google is giving users stronger source preferences across Search and AI results, while Microsoft and Google continue moving paid search toward broader automation.
At the same time, Google is investigating a wave of missing favicons, and a limited Business Profile test is bringing view counts back to Google Posts.
In This Edition
- Google Adds a Preferred Sources Button and Expands Personalization Across Search
- Google Explains Why Some Search Updates Get Announced and Others Do Not
- Microsoft Advertising Is Rolling AI Max Out Globally
- Microsoft Ads Will Remove Max CPC From New Non-Portfolio Campaigns
- Google Is Testing the Return of View Counts for Google Posts
- Google Is Investigating a Spike in Missing Search Favicons
- Google Adds More AI Max Testing and Planning Tools
- Ahrefs Makes the Case for Information Gain Without Calling It a Confirmed Ranking Factor
Google Adds a Preferred Sources Button and Expands Personalization Across Search
Google announced on August 20 that publishers can now embed an interactive Preferred Sources button that lets readers add a site as a preferred source and then returns them to the page they were viewing. Google says content from selected sources is more likely to appear in Top Stories and can receive a preferred badge in AI Overviews and AI Mode for users who chose that source.
Google says more than 600,000 unique sources have already been selected through Preferred Sources. The new embedded button changes the publisher experience because readers no longer have to leave the site and complete the process inside a separate Google preferences page.
Google Search Central’s new documentation says Preferred Sources is available globally for Top Stories in supported Search languages. Preferred sources can also appear in AI Mode and AI Overviews wherever those features are available. Google provides a recommended JavaScript implementation, a customizable implementation, and a deeplink option for publishers that cannot add the interactive button.
The same August 20 announcement also expands personalization outside Search. Google says Discover users will soon be able to describe, in natural language, what they want to see more or less of, while the Google News Android app now allows topic-level customization of daily audio briefings.
Why It Matters
Preferred Sources creates a direct relationship between audience loyalty and personalized search visibility. It is not a universal ranking boost, because the benefit applies to users who explicitly select the source, but publishers and content-heavy brands now have a concrete way to ask loyal readers to influence what they see in Google.
For sites eligible for Preferred Sources, the embedded button is worth evaluating alongside newsletter, social, and repeat-visitor calls to action.
Sources:
https://blog.google/products-and-platforms/products/search/personalize-search-discover-news/
Google Search Central
https://developers.google.com/search/docs/appearance/preferred-sources
Search Engine Journal
Google Explains Why Some Search Updates Get Announced and Others Do Not
Google’s John Mueller said on August 20 that Google runs many systems that update continuously, but it generally announces larger or broader changes when those changes are more likely to create noticeable movement across many sites. Mueller said Google has to balance transparency against creating unnecessary concern around smaller changes.
The explanation adds useful context to the August 2026 spam update covered in yesterday’s Unified SEO Pulse. Google announced that rollout because it represents a broader change to its spam systems, while many smaller ranking-system changes continue without individual public announcements.
Mueller’s explanation also reinforces a distinction that matters during volatile periods: a ranking change does not automatically mean a named Google update occurred. Large announced updates create a useful diagnostic timestamp, but normal Search systems keep changing outside those windows.
Why It Matters
Businesses should use announced update dates as diagnostic markers, not as the only explanation for ranking movement. When Google confirms a broad update, annotate it and compare the timing with site-level changes. Outside those windows, technical changes, competitors, SERP layout shifts, smaller system updates, and demand changes can still move performance.
Source:
Search Engine Roundtable
https://www.seroundtable.com/google-announces-some-updates-41903.html
Microsoft Advertising Is Rolling AI Max Out Globally
Microsoft Advertising is rolling AI Max out globally across advertiser accounts, making the opt-in Search campaign feature broadly available after earlier pilot stages. Microsoft says AI Max can expand query matching beyond an advertiser’s existing keyword list, generate additional text assets, and route users to landing pages that better match their intent.
Microsoft Ads Liaison Navah Hopkins said the global rollout includes brand controls, reporting, and advertiser choice from the start. AI Max remains opt-in for new and existing campaigns.
Microsoft also says advertisers that were already using autogenerated text assets or predictive matching may see those capabilities moved into AI Max settings. Other AI Max features are not automatically activated unless the advertiser opts in.
The broader strategic change is similar to what is happening in Google Ads: Search campaigns are moving away from strict keyword-to-ad-to-URL relationships and toward systems that use page content, contextual signals, and automation to decide which query, message, and landing page combination to serve.
Why It Matters
Advertisers should treat AI Max as a controlled experiment rather than a switch that automatically deserves to be enabled everywhere. Search-term reporting, brand controls, URL rules, conversion quality, and landing-page accuracy become more important when the system can expand beyond the original keyword and URL setup.
Sources:
Search Engine Roundtable
https://www.seroundtable.com/microsoft-advertising-ai-max-41909.html
Microsoft Advertising Help
https://help.ads.microsoft.com/apex/index/22/en/14021
Microsoft Ads Will Remove Max CPC From New Non-Portfolio Campaigns
Microsoft Advertising will stop offering Max CPC when advertisers create new non-portfolio campaigns using several automated bid strategies beginning October 1, 2026, according to Microsoft guidance reported by Search Engine Journal. The affected strategies include Target CPA, Target ROAS, Maximize Conversions, Maximize Conversion Value, and Maximize Clicks.
The change applies only to new campaigns created after the October 1 cutoff. Existing campaigns that already use Max CPC can keep the setting, and Max CPC will remain available with portfolio bid strategies. Target Impression Share and Enhanced CPC are not affected.
Microsoft says a Max CPC ceiling can conflict with automated bidding when a campaign is simultaneously being asked to hit a CPA or ROAS target. The platform is pushing advertisers toward controls tied more directly to business outcomes, including budgets, target CPA or ROAS settings, conversion value rules, and seasonality adjustments.
This is not an immediate migration requirement. Existing campaigns do not need to remove their Max CPC limits solely because the setup option is disappearing from new campaigns.
Why It Matters
Teams that use Max CPC as a safety control should review campaign templates before October 1. More importantly, the change raises the cost of weak conversion tracking. As manual bid constraints disappear, automated bidding becomes more dependent on accurate goals, conversion values, and realistic performance targets.
Source:
Search Engine Journal
https://www.searchenginejournal.com/microsoft-ads-is-removing-max-cpc-from-new-campaigns/586527/
Google Is Testing the Return of View Counts for Google Posts
Google appears to be testing a new “people viewed” metric on Google Business Profile Posts, according to screenshots reported by Search Engine Roundtable on August 20. The metric resembles the Google Posts Insights data Google originally launched in 2018 and discontinued in 2023.
The current test appears limited. Search Engine Roundtable’s Barry Schwartz said he could not reproduce the view count in his own interface, and the original report came from a screenshot shared by another user.
Google recently began surfacing view counts for some Business Profile photos and videos as well, which makes the Posts experiment part of a broader return of lightweight visibility metrics inside the local business interface.
The test currently shows views rather than a full restoration of the older Posts Insights reporting. There is no confirmation that Google will restore click reporting or roll the feature out to every Business Profile.
Why It Matters
Google Posts have been difficult to evaluate because businesses can publish updates without much native performance feedback. Even a basic view count would make it easier to compare topics, offers, events, and update formats over time.
For now, local businesses should treat this as a limited test and avoid rebuilding reporting around it until Google confirms broader availability.
Source:
Search Engine Roundtable
https://www.seroundtable.com/google-posts-insights-views-41913.html
Google Is Investigating a Spike in Missing Search Favicons
Site owners have reported a recent spike in favicons disappearing from Google Search result snippets, with affected listings showing Google’s generic globe icon instead. Search Engine Roundtable collected multiple complaints from site owners whose favicon files remained accessible, correctly referenced, and unblocked.
The complaints began before the August 2026 spam update, so there is no evidence tying the favicon problem to that rollout. Search Engine Roundtable also reported that at least one affected example began working again after publication.
Google Search executive Rajan Patel responded on August 20 that Google is looking into the issue. Google has not yet published a formal incident notice or explanation.
Because the reports include sites whose favicon files appear technically valid, the pattern may involve Google’s processing or display systems rather than a single shared implementation error. That remains unconfirmed until Google provides more detail.
Why It Matters
A missing favicon is a presentation problem, not proof of a ranking penalty. If a site’s icon disappears, first confirm that the favicon still follows Google’s technical requirements and is crawlable. If the implementation is sound, avoid unnecessary site changes while Google investigates the wider reports.
Source:
Search Engine Roundtable
https://www.seroundtable.com/google-search-favicons-breaking-41904.html
Also Worth Watching
Google Adds More AI Max Testing and Planning Tools
Google announced new AI Max experimentation and forecasting tools on August 20. Starting in September, advertisers will be able to test different budgets and ROI targets across multiple Search campaigns in one A/B test. Google is also allowing AI Max experiments to run while brand and location controls remain enabled.
Performance Planner is gaining additional forecasting for bidding and budget changes, with the option to apply suggested changes directly to campaigns.
The important part is not another layer of automation by itself. Google is adding more tools for testing automation before committing larger budgets, which gives advertisers a better opportunity to separate incremental gains from changes they would have received anyway.
Sources:
https://blog.google/products/ads-commerce/ai-max-testing-planning-tools/
Search Engine Land
https://searchengineland.com/google-adds-new-ai-max-testing-and-planning-tools-485566
Ahrefs Makes the Case for Information Gain Without Calling It a Confirmed Ranking Factor
Ahrefs published a detailed August 20 guide to “information gain,” defined as the additional useful information a page contributes beyond what competing pages already provide. The article draws heavily from a Google patent but explicitly notes that Google has never confirmed a live information-gain score as part of its ranking or AI systems.
That distinction is important. The useful strategy is not to claim a hidden ranking factor has been discovered. It is to identify what the existing SERP already says, then add original data, first-hand experience, examples, or perspectives that genuinely expand the answer.
For businesses publishing increasingly similar AI-assisted content, that is a practical differentiation test even without a confirmed algorithmic score.
Source:
Ahrefs
https://ahrefs.com/blog/information-gain/
The Unified Take
Today’s stories are all variations on a shift from direct control toward guided systems.
Google’s Preferred Sources feature gives users more control over which publishers Google highlights, but only after the user explicitly chooses them. Microsoft AI Max and Google AI Max give advertising systems more freedom to interpret queries, assets, and landing pages, while advertisers are increasingly asked to control outcomes through goals, guardrails, and experiments rather than exact matching rules. Microsoft’s Max CPC change pushes that same direction further by removing one manual constraint from new automated campaigns.
At the same time, the favicon issue and Google Posts test show why marketers still need to distinguish platform behavior from website behavior. A missing favicon can look like a site problem even when Google may be investigating a broader display issue. A new Business Profile metric can look like a permanent reporting feature when it is still only a limited test.
The practical pattern is consistent: know which controls belong to you, which belong to the platform, and which signals are still experimental.
For organic search, that means strong technical accessibility, recognizable source identity, and content that adds something competitors do not. For paid search, it means better conversion inputs and disciplined experimentation as platforms automate more of the campaign. The systems are becoming more adaptive, but that makes measurement and clear decision rules more important, not less.
