With a background as a global leader in SEO, content marketing, and data analytics, Anastasia Braitsik is a leading voice navigating the complexities of modern digital marketing. As Google Ads undergoes a seismic shift, moving from a keyword-driven system to an AI-powered, intent-based auction, many advertisers are finding their trusted playbooks obsolete. Today, we sit down with Anastasia to unpack this new reality. We’ll explore how to adapt campaign structures, rethink landing page strategies, and measure success when the user’s “why” matters more than the words they type. This conversation will delve into the practical changes needed to thrive in an era where AI infers user needs and reshapes the entire advertising landscape.
Google’s auction is now triggered by inferred intent, not just keywords. Using a scenario like a user asking, “Why is my pool green?”, could you break down how the AI infers commercial intent and what practical steps an advertiser for pool supplies should take to appear?
It’s a fantastic and very real example of the deep changes happening under the hood. When a user asks, “Why is my pool green?”, they’re clearly in troubleshooting mode, not shopping mode. In the old system, an advertiser for pool supplies would likely never appear for such an informational query. But now, Google’s AI uses a process we can think of as a “query fan out.” It instantly deconstructs the question into multiple subtopics, running concurrent searches to build a holistic answer. The crucial part is its reasoning layer; it doesn’t just see a question, it sees a problem that a product can solve. The AI infers that a green pool will ultimately require a chemical solution, so it preemptively triggers an auction for pool-cleaning supplies and serves those ads alongside the organic explanation. For an advertiser, the practical step is to stop fixating only on bottom-of-the-funnel keywords. You need to build campaigns and landing pages that address the user’s problem state. Your ad copy and on-page content should speak directly to solving the “green pool” issue, positioning your product as the hero in that story. You’re no longer just matching a keyword; you’re matching your offering to an inferred need.
For a team shifting from a keyword-first to an intent-first model, how does campaign structure fundamentally change? Please walk us through how you would group keywords and write ad copy for a user with a “feature comparison” intent versus one who is “ready to buy.”
This is the core of the mental model shift. You stop treating keywords as the organizing principle and start mapping everything to the why behind the search. Let’s take the query “best CRM.” This could reflect two very different intents. For a user in the “feature comparison” stage, I would group keywords like “compare CRM platforms,” “best CRM for small business,” or “CRM features list.” The ad copy would speak directly to that exploratory goal, maybe with a headline like “Find Your Perfect CRM in Minutes” and descriptions that highlight comparison tools, feature breakdowns, or free guides. The landing page would then deliver on that promise with a detailed comparison chart, not a hard “buy now” pitch. In contrast, for a user who is “ready to buy,” the intent is validation. They might still search “best CRM,” but the surrounding context signals they’re closer to a decision. For this group, your ad copy needs to be more assertive: “The #1 Rated CRM” or “Start Your Free Trial Today.” You’re speaking to their goal of making a final choice and reducing risk, so you’d group keywords that show that final-stage thinking and direct them to a page that facilitates a quick and confident purchase or sign-up.
To appear in new AI-driven formats, campaigns often need broad match or Performance Max, yet exact match remains useful for brand defense. Could you elaborate on this strategic trade-off? What performance shifts and measurement adjustments should advertisers anticipate when embracing broader targeting for this new visibility?
This is a critical balancing act every advertiser faces now. If you want to be eligible for placement within AI Overviews or the new conversational AI Mode, you simply have to embrace broader targeting. Exact and phrase match just won’t get you into that exploratory layer where the system is reasoning through answers. So, you must use broad match keywords or run campaigns like Performance Max. However, exact match is still your ironclad defense for your own brand terms and high-visibility, high-intent searches where you want total control above the AI summaries. The trade-off is control for reach. When you embrace broader targeting, you must adjust your performance expectations. This traffic is, by nature, higher up the funnel. You’ll see conversion rates that don’t match your branded, bottom-of-the-funnel searches. That’s okay, as long as you plan for it. The measurement adjustment is to stop chasing immediate ROAS from these placements and instead look at how they contribute to the entire funnel, even if you can’t perfectly isolate their performance since Google doesn’t segment reporting for these AI-specific formats.
The system now rewards deep contextual alignment between an ad and a landing page. Beyond just rewriting copy to solve a “why,” what specific on-page elements, asset types, and first-party data signals are most crucial for training the algorithm to win these intent-based auctions?
Contextual alignment is everything. The AI is building an answer to a problem, and if your landing page directly addresses that problem, you’re far more likely to win the auction. It’s not just about rewriting a headline. It’s about a holistic approach. The algorithm prioritizes rich metadata, so your page titles and descriptions must be dialed in. It also heavily favors multiple high-quality images and video assets that visually demonstrate the solution. For e-commerce, an optimized shopping feed with every single relevant attribute filled out is non-negotiable. These assets give the AI more data points to confirm your relevance. Critically, first-party data is your superpower. By uploading Customer Match lists, you are explicitly teaching the AI what your highest-value customers look like. This training directly affects its bidding behavior; the system becomes more aggressive and smarter when it sees a user who mirrors the profile of your best customers. This creates a powerful feedback loop that goes far beyond simple copy changes.
AI-powered campaigns often need significant conversion data to learn, creating a potential “scissors gap” for smaller advertisers. Given the lack of specific reporting for AI placements, what alternative strategies or proxy metrics can a business with a limited budget use to compete effectively?
This “scissors gap” is a very real challenge. AI-powered campaigns like Performance Max often need a minimum threshold, say 30 conversions in 30 days, to even begin learning effectively. For smaller businesses with lower budgets or longer sales cycles, hitting that volume is tough. Since Google doesn’t give us specific reporting for AI placements, you can’t just throw money at it and see what works. Instead of focusing solely on the final sale, smaller advertisers should use proxy metrics. Track micro-conversions like newsletter sign-ups, guide downloads, or video views as signals of engagement. These can provide the algorithm with some positive data to learn from, even if it’s not a final purchase. Another strategy is to focus intensely on a very specific niche where you can dominate the intent signals. Instead of competing broadly, own the conversation around a very particular problem your product solves. This allows you to generate more concentrated data within a smaller budget, helping you train the algorithm and close that gap without having to outspend massive competitors.
What is your forecast for Google Ads?
My forecast is that the line between search and discovery will completely dissolve. We’re moving away from a system where you type a query and get ten blue links, toward a continuous, conversational dialogue with an AI that anticipates your next need. For advertisers, this means the concept of a “campaign” as a static set of keywords and ads will become a relic. The future is about providing the system with a rich portfolio of business goals, creative assets, and first-party data, and then trusting the AI to assemble the right message for the right user in the right format at the right moment. The most successful advertisers won’t be the best keyword pickers; they’ll be the best teachers, continuously feeding the AI the signals it needs to understand their business and customers on a deeply contextual level. It’s a fundamental shift from manually managing inputs to strategically guiding an automated system.
