Anastasia Braitsik is a global leader in SEO, content marketing, and data analytics who has spent years navigating the volatile shifts of the digital advertising landscape. As a prominent expert in paid search strategy, she specializes in helping brands maintain profitability during periods of rapid technological disruption. Today, she joins us to discuss the seismic shift caused by the integration of AI Overviews into the search experience, offering a roadmap for advertisers to protect their margins and evolve their tactics.
The following discussion explores the statistical reality of declining click-through rates in high-penetration sectors like finance and retail, where AI now handles a vast majority of long-tail queries. We delve into the phenomenon of the collapsing buyer journey, the strategic distinction between informational and transactional keyword value, and the technical rigors required for modern product feed optimization. Finally, we examine the pivot toward first-party data and unique value propositions as the primary levers for success in an increasingly automated bidding environment.
In sectors like finance and retail, AI summaries now appear on roughly 80% of long-tail searches, causing paid click-through rates to drop significantly. How are you adjusting bidding for these high-penetration categories, and what specific metrics should teams prioritize to gauge account health beyond traditional click volume?
The shift is undeniable; in finance, we see AI Overviews on 79% of longer queries, and that number climbs to 84% for retail comparison searches. When paid CTR plummets by 68% as it did recently, dropping from nearly 20% to just 6.34%, we have to stop chasing raw click volume as a primary KPI. We are adjusting bidding by moving away from aggressive manual CPCs on informational terms and allowing automated bidding systems to optimize toward conversion predictions rather than just traffic. Instead of looking at clicks, teams must prioritize conversion rate trends and “effective CPA” to see if the higher intent of remaining clicks offsets the rising costs. I also keep a very close eye on impression share for ads appearing specifically above the AI Overview, as those positions still perform reasonably well compared to the ghost town below the fold.
Buyer journeys are currently collapsing from weeks into minutes as automated summaries handle the research phase, leading to higher conversion rates in industries like education. How can advertisers capitalize on this pre-qualification, and what steps prevent overpaying for these clicks amidst rising cost-per-click inflation?
It is fascinating to see industries like education and instruction experience a 43.87% jump in conversion rates because the AI has already done the heavy lifting of answering the user’s basic questions. To capitalize on this, your landing pages shouldn’t repeat the basic definitions the AI just provided; they need to move straight to the “apply now” or “get a quote” phase. To prevent overpaying, we use value-based bidding to ensure we aren’t just throwing money at a shrinking pool of clicks, especially since Google Search spending grew 9% while click growth was only 4%. We balance the 45% CPC inflation by monitoring the total return on ad spend, acknowledging that while we pay more per click, the 7% conversion rate we see now is much more valuable than the 5% we saw in a pre-AI world. It’s about accepting the quality-over-quantity trade-off and tightening the belt on broad, non-converting traffic.
Informational queries face heavy cannibalization while transactional terms remain resilient. When auditing several months of query data, how do you distinguish between “how-to” terms that still deliver value and those that are budget drains? Please provide a step-by-step breakdown of your keyword evaluation process.
I start by pulling at least 90 days of Google Ads query data and flagging anything containing “how to,” “what is,” or “guide.” Then, I cross-reference this with Google Search Console to see which of these terms are triggering AI Overviews, as GSC now tags these specifically in performance reports. The next step is a cold, hard look at the numbers: if a query shows less than a 1% CTR and contributes less than half of my account’s average conversion rate, it is likely a budget drain. However, there is a major exception if our brand is actually cited within the AI Overview itself, which can lead to a 91% lift in paid CTR. For the losers, I either slash the bids to maintain a low-cost presence or shift that budget entirely into transactional terms like “buy” or “near me” where the user still needs to click away from the AI to finish their task.
The current shopping environment relies on billions of product listings refreshed hourly, prioritizing specific attributes and rich media. What are the technical requirements for optimizing a product feed to ensure visibility, and how does providing five or more images or video content change performance in AI-driven results?
Google’s Shopping Graph is a beast with 50 billion listings, and if your data isn’t refreshed hourly, you risk being filtered out of the AI Mode entirely. Technical optimization now requires extreme attribute specificity—think natural language terms like “breathable bamboo” or “waterproof for rainy commutes”—because the AI searches for these exact matches to generate its summaries. We have observed that the AI prioritizes listings with five or more product images and integrated video content, as these rich media assets are now essential for driving visual discovery and virtual try-ons. Providing these assets transforms your performance by making your product the “featured” recommendation in an AI comparison table rather than just a text link at the bottom. It turns a static listing into a dynamic recommendation that the AI feels confident presenting to a high-intent shopper.
Generic headlines are losing effectiveness because users now click ads after reading AI-generated comparisons. How should creative messaging pivot to highlight unique offers like “free migration” or “risk reversal,” and what role does first-party audience data play in reaching users when keyword-based targeting becomes less reliable?
You can no longer afford to be generic; a headline like “Tax Preparation Services” is dead weight compared to “Same-Day CPA Review | $50 Off.” Since the user has already read the AI’s summary of what your product does, your ad must explain why they should choose you right this second, using “risk reversal” offers like free trials or no-credit-card-required models. This is where first-party data becomes our secret weapon; we use Customer Match lists—now accessible with as few as 100 users—to target people who have already interacted with us. By feeding our CRM data into Performance Max and Demand Gen campaigns, we allow Google’s AI to find “lookalikes” of our best customers based on actual purchase history. This ensures our ads are shown to people who are ready to convert, even when the specific keywords they use are being swallowed up by the AI’s informational summaries.
What is your forecast for the future of paid search?
I believe paid search will evolve into an “audience-first, keyword-second” discipline where the most successful advertisers are those who own their data. We are going to see a total move away from old-school, keyword-heavy campaigns in favor of formats like Performance Max that prioritize first-party data and rich media assets. The “winner-take-all” dynamic of AI citations will intensify, making it critical for brands to not only bid on terms but to ensure their site content is authoritative enough to be cited by the AI. Ultimately, search will become less about being the first blue link and more about being the most trusted recommendation within a seamless, AI-curated conversation. Success will depend on your ability to feed the machine high-quality signals and offer unique, differentiated value that an AI summary simply cannot replicate.
