Digital Marketing Updates: July 2026

Paid Media

Updates by Casey Anderson, Paid Media Director

OpenAI Users Adapting to Ads with 50% Fewer Dismissals

Since introducing ads to the free tiers of ChatGPT, OpenAI is reporting more on user patterns. The most notable trend that has occurred in this short amount of time is users’ ad dismissal rate. What started as a relatively high dismissal rate when ads were integrated, has decreased by 50%.

Why is this important? It’s giving both OpenAI and advertisers an indication of how receptive users are to this new type of ad. Seeing that fewer users are dismissing ads also tells us more about ad relevance aligning with interest and intent.

Given that this is still a new platform and advertisers are learning how to make their ads relevant to the ChatGPT audience, this helps to guide our understanding of best practices in an environment where there is yet to be any fully established.

Interested in running ChatGPT Ads? Talk to an expert on our paid media team about a strategy assessment.

Expansion of B2B Targeting on Microsoft Ads

Microsoft Ads has had a leg up on Google with its LinkedIn-fed audience targeting options that were released more than a year ago. An expansion of this targeting is now covering 10 job seniority levels including CXO, VP, Director, Manager, Senior, Entry, Owner, Partner, Training, and Volunteer.

This update allows B2B advertisers to target or observe these LinkedIn audience options at the campaign or ad group level, providing a more advanced way to align goals with messaging, bid strategies and reporting.

Google Ads Bidding Changes – What to Know

If you manage campaigns often limited by budget, news about backend bidding optimizations is likely to affect you. The change, expected to take place on August 17, will make CPA and ROAS targets more predictable and help campaigns limited by budget align performance with those targets, essentially promising smarter smart bidding.

A common practice in value-based bidding has been setting CPA and ROAS targets above actual metrics in order to ensure the conversions could still be found within a target CPA or target ROAS range. Sometimes this caused volatility when additional budget was added to budget-constrained campaigns. If targets were set too low and conversions couldn’t be found at those targets, campaigns would stop serving completely.

A notification about this is expected to appear in accounts around July 6. According to Search Engine Journal, “the July 6 account notifications are particularly important for teams managing budget-limited campaigns. Google’s recommendation to review CPA and ROAS targets suggests that some advertisers may discover their existing targets no longer reflect current business conditions or business goals.”

SEO & AEO

Updates by Brett Woodward, Sr. SEO

Google adds generative AI impressions to Search Console

For the last several years, Google has lumped all data about how a site shows up in AI Overviews and AI Mode into the regular search performance report. 

The data was there, Google assured us, but there was no way to separate out performance on traditional search versus AI answers.

That is finally changing, with Google beginning to roll out a Generative AI report to Search Console accounts.  

In the Generative AI report you’ll be able to see impression figures for your site’s pages…and that’s it. Unfortunately you won’t find a list of queries your site is generating AI visibility for or a numerical representation of how much traffic those AI impressions drove. For now, we’ll have to be content with what Google gives us.

It’s important to note that in this report impressions mean an instance where a link to your site is shown in an AI response. Unlinked mentions are not counted here, and links that only appear when a user expands an AI answer are only counted if the user performs that action. Google Search liaison John Mueller explained this on X:

While I don’t think we should be holding our breath for query and click data to appear in this report any time soon, we can get to work analyzing what pages are being shown to users in AI responses. 

Specifically, we’re digging into the differences between top pages in traditional search results with the top pages in this Generative AI report. 

This report is in the process of rolling out to accounts. We’re seeing it appear for larger sites initially, and it may be a few weeks before all accounts get this feature.

Study finds AI avoids “commodity content”

An interesting study by Flying V Group and GEO Genius indicates that content that’s easy for LLMs to summarize with existing training data is less likely to be included in an AI response than content that offers unique insights and information.

The study measured content across four large online publishers and found that the pages that earned the most visibility in AI contained more information that was likely not part of a model’s training data. 

Conceptually this makes perfect sense. AI will lean on its training data to provide an initial response for a query, and there is greater value in citing sources that provide new, supplemental information than sources that largely repeat what the LLM already knows.

The study hinges on commodity content scores, which is a metric that GEO Genius built a specific tool to measure. According to the Flying V Group, the tool “analyzes any URL and returns a score showing how saturated, replicated, and well-trodden a topic has become”. The more the content references a heavily covered topic without adding unique information or perspective, the higher the commodity content score.

If you’re interested in the tool, join the waitlist as they expand access to more users.