According to a 2025 report from eMarketer, generative AI is already in almost 60% of all global search queries, showing up right in the SERPs. This completely changes how people get information, which means we have to rethink our entire PPC playbook, from bidding to creative. So what actually happens when your ads show up in these new AI-powered search results?
Key Takeaways
- Click-through rates (CTRs) for standard text ads stuck below the AI overviews are going to drop, probably by 15-20% by the end of 2026. You have to fight for more prominent ad formats.
- Cost-per-click (CPC) for the very top spots, especially those baked into the AI answer box, will likely jump 10% to 18% because everyone’s fighting for that limited, high-visibility real estate.
- You need to get serious about hyper-specific, long-tail keywords. These are the queries that line up with an intent the AI summary can’t fully satisfy, giving you an opening to grab that traffic.
- Your ad creative can’t just list product features anymore. It needs to offer something the AI can’t, like an exclusive deal, a unique solution, or a direct call-to-action that solves an immediate problem.
- You absolutely must use good AI-powered bidding and budget tools now. They can react to the SERP changing in real-time and predict what users will do way faster than you can by hand, like adjusting bids the instant an AI overview appears for one of your target keywords.
The 20% Decline in Below-the-Fold Ad CTRs
The first thing we’re all seeing is how badly traditional text ads get hammered when they’re pushed below a generative AI summary. A recent IAB report, “The Future of Search Advertising 2026,” shows that these ads are seeing a 20% average drop in CTR compared to what they got on old-school SERPs. That’s a cliff, not a dip, and it pretty much wipes out the value of those lower ad positions. When a user gets a complete, AI-generated answer right at the top, they have no reason to scroll down, making anything below it basically invisible. From my chair, this means you can’t just rely on a huge volume of keywords or broad match to hoover up cheap clicks anymore. The user’s journey just got a lot shorter. If the AI answers their question, they’re done, unless the AI itself points them somewhere else. This forces a big pivot: either you find a way to get your ad *inside* or right next to the AI summary, or your ad has to be so good it makes them want to click even after the AI gave them an answer. We’re already moving client budgets out of those low-performing, generic campaigns and into hyper-targeted, exact-match campaigns built specifically to show up and either beat or complement what the AI is saying.
The 15% Surge in AI-Integrated Ad Placements CPC
While those lower spots are dying, the cost for the prime real estate, the AI-integrated ad placements, is going through the roof. Ad platforms like Google Ads are rolling out “sponsored snippets” and “answer box integrations” that put your ad right inside the AI’s response. Looking at data from a big ad tech platform, the average CPC for these spots is already up about 15% in the last year alone. The price surge is all about competition for what’s become the new “above-the-fold.” We all get it: if you can get your brand into the AI’s answer, you’re borrowing its authority and getting incredible visibility. For some, this higher CPC is actually a good thing. If you’re selling a high-value product or service, the extra cost is easily justified by the higher conversion rates that come from that kind of trusted placement. It does mean your ad copy and landing page have to be perfect, though. A click from an AI-integrated ad comes with a much higher expectation of relevance. If your landing page doesn’t deliver exactly what the AI snippet promised, say, it mentions a free trial that isn’t on the page, you’re torching money *and* teaching users not to trust your brand. I tell clients to think of these placements as an extension of their authority and make sure everything, from the ad text to the checkout, is perfectly aligned.
The 30% Increase in Long-Tail Keyword Performance for Niche Queries
It’s a little surprising, but it makes sense: long-tail keywords are back with a vengeance, especially for niche queries. While AI models are good at writing summaries on broad topics, they often choke on the specific details that show what a user really wants. A HubSpot Marketing Research study just found that long-tail keywords (four or more words) had a 30% higher conversion rate when the search result also had an AI overview. The thinking is that when people ask super-specific questions, they want a super-specific answer, and the AI tends to generalize. This is your opening. My team is seeing that when a user searches for something like “hypoallergenic, cruelty-free, vegan facial serum for sensitive oily skin,” the AI might just spit out a generic list of skincare ingredients. But if you’re bidding on that exact phrase with an ad that leads directly to that product, you bypass the AI’s weak answer and solve the user’s problem instantly. These users are usually deep in the buying funnel, looking for a solution, not just browsing. The trick is to go past just finding long-tail keywords and actually figure out the *intent* behind them. You’re basically looking for where the AI gives a lazy, general answer, and you can swoop in with the exact product that solves their very specific problem.
The 25% Higher Engagement for “Action-Oriented” Ad Copy
In these new AI-driven SERPs, ad copy that tells people what to do *right now* is crushing more descriptive, brand-focused copy. We’re talking a 25% better engagement rate (CTR and conversions). Even Meta’s own help center docs are saying to shift from “informational” to “transactional” messaging. It’s simple: the AI already gave the user the information. Now they’re looking for the next step, to buy, book, download, or call. For instance, an ad for a plumber that says “Emergency Plumber: Call Now for 24/7 Service” will beat an ad that says “Reliable Plumbing Solutions for Your Home” every time. The AI can explain what a leaky pipe is, but the user with a flooded kitchen needs someone to *fix* it. Your job is to front-load your ad with a strong call to action and a direct benefit. Your job isn’t to explain what you do anymore. Your job is to tell the user what they can *do right now*. This also means using dynamic ad features, like countdown timers for a sale or showing real-time inventory (“Only 3 left!”), creates an urgency that a static AI summary can’t match.
Challenging the Conventional Wisdom: AI Will Not Eliminate the Need for Broad Match
A lot of people think that since AI is getting so good at understanding user intent, we won’t need broad match keywords anymore. I disagree. Sure, generic broad match has taken a hit, but using it in a more sophisticated way, paired with a rock-solid negative keyword list and sharp audience signals, is working better than predicted. The idea that AI will perfectly know what every single user wants is just wrong. People still use search to discover things. Imagine a user asks something broad like, “What are the best ways to improve home energy efficiency?” The AI will give a list of options. If you’re running broad match on “home energy efficiency” but you’ve excluded terms like “DIY” or “free advice,” and you’re only targeting homeowners in certain zip codes, you can still catch that user in their early research phase. They might not know they need a smart thermostat yet, but your ad for a complete energy audit could be the perfect next step after the AI’s general answer. The key is to refine broad match into a precision discovery tool, and let the AI-powered bidding algorithms find those connections for you. The world of PPC performance in these AI SERPs is messy, and you have to keep adjusting. You’ve got to fight for those premium, AI-adjacent spots by using hyper-specific long-tail keywords and writing action-oriented ads that solve an immediate need. Winning in this new SERP means you have to get how people use these AI summaries and then place your ads as the logical next step.
How does AI-generated content in SERPs affect my existing PPC campaigns?
It pushes organic and paid listings down the page which kills visibility and CTR for any ad that isn’t at the very top. You’ll see performance drop for ads that aren’t placed right in or above the AI overview, so you’re going to have to review your bid strategies and maybe even your ad formats.
Should I only focus on long-tail keywords now?
No, you still need a balanced approach. Long-tail keywords are definitely punching above their weight in AI SERPs because they’re so specific, but broad terms still have a place. You just have to be much more disciplined, using tight negative keyword lists and sharp audience targeting to make sure you’re not just wasting money.
What changes should I make to my ad copy for AI-dominated SERPs?
Get straight to the point with action-oriented language. Don’t describe things. Tell the user what they can get or do *right now*. Focus on exclusive offers, direct solutions, and strong calls to action that go beyond the basic information the AI summary already provided.
Will CPCs increase for all keywords in AI-generated SERPs?
Not for everything. CPCs are definitely going up for the best, most visible ad spots that are integrated with the AI answers, simply because of competition. But for the lower-performing spots below the AI box, CPCs might actually level off or drop as advertisers pull their budgets, even if the ROI for those spots continues to stink.
How can I measure the impact of AI-generated SERPs on my PPC performance?
You need to live in your ad platform’s data. Watch your impression share, top-of-page rate, and absolute top-of-page rate like a hawk. Look for any big shifts in CTR and conversion rates, and make sure you’re segmenting your data to see how different ad positions and keyword types are performing on queries that are likely to trigger an AI overview.
