Generative AI in search has totally upended how people find things, and it’s causing some major shifts in organic traffic. New AI search studies keep pointing to a “zero traffic” problem, where users get their answers right on the results page and never click through to a website. This case study breaks down a recent campaign we ran for a B2B SaaS provider that was getting hit by this, trying to see if we could reclaim some of that visibility by adapting our content strategy.
Key Takeaways
- Re-optimizing existing content for direct answer boxes and featured snippets can increase SERP visibility by up to 35% even with declining click-through rates.
- Implementing a strong schema markup strategy, particularly for FAQs and how-to guides, directly contributed to a 15% increase in branded search queries.
- Focusing on long-tail, conversational keywords saw a 20% improvement in impression share within AI-generated summaries compared to broad terms.
- A/B testing different call-to-action placements within content designed for zero-click environments revealed that a mid-article, benefit-driven CTA yielded 8% more conversions than traditional end-of-page placements.
- Analyzing AI search result snippets for competitor language and gaps allowed for the creation of differentiated content that captured a 10% higher engagement rate in subsequent user tests.
Campaign Overview: Adapting to the AI Search Frontier
Our client, a supply chain software vendor called “LogiFlow Solutions,” started seeing the warning signs of AI’s effect on search back in late 2025. Their organic traffic, which had always been a huge source of leads, took a worrying 12% dip over two quarters even though their SEO efforts were consistent. Their top-ranking pages were simply seeing fewer clicks, pointing to a problem deeper than just rankings. So, we kicked off a specialized “AI Search Adaptation” campaign that ran from January to April 2026 with a budget of $45,000.
Our main goal was to grow, or at least maintain, qualified leads from organic search by tackling the zero-click trend head-on. The plan was to get LogiFlow’s name and content inside AI-generated summaries, featured snippets, and “People Also Ask” boxes, grabbing user attention even without a click. To call it a success, we needed to keep our Cost Per Lead (CPL) under $150, hit a 5% bump in branded search volume, and see a modest 2% growth in qualified demo requests coming from organic.
“In 2025, nearly a third of Gen Z consumers (31%) defaulted to chatbots or AI to find information online, according to HubSpot’s 2025 AI Trends for Marketers report, and it’s clear that AI search is growing, rapidly.”
Initial Strategy: Content Reframing and Schema Enhancement
Our strategy boiled down to two things: content reframing and advanced schema markup implementation. Our theory was simple: by structuring content to give concise, direct answers to common questions, it would be much more digestible for the AI models putting together search summaries. At the same time, strong schema would explicitly tell search engines what the content was about and highlight the key data points for them.
We started with an exhaustive audit of LogiFlow’s top 50 organic landing pages, flagging the ones that ranked for informational queries AI was likely to answer directly. A page on “How to optimize warehouse layout for efficiency,” for example, was a perfect candidate. We then analyzed the live AI search results for these queries, taking notes on the exact language used in snippets and what kinds of questions the AI overviews were answering.
With that intel, our team went in and revised the pages. Every targeted page got a clear, tight summary right at the top (usually 50-70 words) written specifically to be an ideal featured snippet or AI summary. We broke down complex subjects into simple Q&A sections with headings like “What is X?” and “How does Y impact Z?”. This meant rewriting whole paragraphs, adding bulleted lists, and making sure key terms were bolded. On that warehouse optimization page, for instance, we added a new section called “Key Principles of Efficient Warehouse Layout” with five bullet points that each answered a common follow-up query.
While the content team was busy, the technical team implemented extensive structured data. We used FAQPage schema on all the new Q&A sections, HowTo schema for any step-by-step guides, and beefed up the Article schema with more granular properties like about and mentions to give Google more context. This involved carefully mapping content elements to the right schema types and then running everything through Google’s Rich Results Test to make sure it was working.
Creative Approach: Precision and Persuasion in Micro-Content
Our creative strategy shifted away from broad, keyword-stuffed articles toward hyper-focused, answer-oriented micro-content blocks. The goal was to become the definitive, concise answer that AI would grab, moving beyond just ranking for the term. We adopted a “snippet-first” mentality. We started crafting headlines as direct questions and then immediately followed them with a tight answer in the first paragraph, often under 30 words. So instead of a heading like “The Benefits of Inventory Management Software,” we changed it to “What are the core benefits of inventory management software?” and put the answer right below it.
Visuals had a role to play, too. While AI search is mostly about text, we noticed that pages with relevant, high-quality infographics that explained a process were sometimes getting pulled into visual snippets, which can still drive some traffic. We whipped up five new infographics for key supply chain concepts, embedded them on the right pages, and made sure their alt text and captions were descriptive.
We also got experimental with call-to-action (CTA) placement. Since users getting an answer at the top might not scroll, we started integrating subtle, contextually relevant CTAs much earlier in the text. For example, right after explaining a specific software feature, a small banner might pop up saying, “See how LogiFlow handles [feature] in a live demo,” linking straight to the demo form. This was completely different from our standard practice of sticking CTAs at the bottom of the page.
Targeting and Measurement: Beyond Traditional Keywords
Our targeting had to evolve past old-school keyword research. We used tools like Ahrefs and Semrush to find what we called “Pillars of Information” (POI), queries that kept popping up in AI summaries and People Also Ask boxes. These were often longer and more conversational than our typical target keywords. Instead of just “inventory control,” we started targeting full phrases like “how does predictive analytics improve inventory control accuracy?”
We measured success from multiple angles. While we still tracked traditional metrics like impressions, clicks, and average position, we put a heavy emphasis on new indicators. We lived in Google Search Console, monitoring changes in featured snippet impressions and clicks via the “Search Appearance” report. We also paid for third-party rank trackers that could specifically tell us if our content was showing up in AI Overviews. And of course, we watched branded search volume in Google Analytics, looking for a lift in searches for “LogiFlow.”
Conversion tracking had to be tight. We used distinct UTM parameters for the new CTAs embedded in our AI-optimized content versus the standard ones, which let us accurately attribute demo requests and downloads. Our CRM was also plugged in so we could track the entire lead journey, from that first organic touchpoint all the way to a qualified lead.
What Worked: Surprising Gains in Visibility and Engagement
The campaign delivered some definite wins. Our work on featured snippets, in particular, paid off. In the first two months, featured snippet impressions shot up by 38% across the pages we targeted. Even with a modest 2.5% average click-through rate (CTR) for these snippets, the jump in raw visibility was huge.
Branded search queries were our biggest win. By consistently showing up in AI summaries for broad informational queries, users just became more aware of LogiFlow Solutions. We saw a 10% increase in branded search volume quarter-over-quarter, which blew past our 5% target. It shows that repeated exposure in AI answers builds brand recall, even if you don’t get the immediate click.
Those early-placement CTAs also did much better than we anticipated. On the AI-optimized content, the mid-article, contextual CTAs had a conversion rate of 1.8%, compared to just 0.9% for the old end-of-article CTAs on similar pages. It gave users who got their answer a clear next step to explore the actual solution.
Our final CPL for organic leads from these pages came in at $135, well under our $150 goal, which told us these leads were good quality. The overall Return on Ad Spend (ROAS) for our organic channels, once you factor in the campaign cost, actually ticked up by 3% from the previous quarter, stopping the decline we’d been seeing.
Key Campaign Metrics (Jan-Apr 2026)
- Budget: $45,000
- Duration: 4 Months
- Total Impressions (Targeted Pages): 1,850,000
- Total Clicks (Targeted Pages): 42,550
- Overall CTR (Targeted Pages): 2.3%
- Total Conversions (Demo Requests): 315
- Cost Per Lead (CPL): $135
- Branded Search Volume Increase: 10%
- Featured Snippet Impression Increase: 38%
But not everything worked. The unpredictable nature of AI search results was a constant headache. Even with our best efforts, some of our most optimized pages would suddenly lose a snippet to a competitor who was less authoritative but maybe had one sentence that was worded more concisely. It just goes to show that while good structure gets you in the game, the AI’s final interpretation of intent and relevance can change day-to-day, which means your snippet position can vanish overnight.
We also ran into some keyword cannibalization. In our rush to create specific Q&A sections, some of the new content started competing with our existing, broader pages for the same terms. For example, a new FAQ bit on “inventory forecasting models” began outranking our much deeper “Complete Guide to Inventory Forecasting” for some long-tail queries which was exactly what we didn’t want. That forced us to go back and do a lot of careful internal linking clean-up and even merge some content.
Direct traffic from the AI overviews themselves was also a letdown. While our impressions went up, the direct click-through rate from those super-summarized results was always really low, often under 1%. It confirms that AI is there to spit out immediate answers. Getting someone to click requires giving them a very good reason to dig deeper.
Optimization Steps: Iteration and Refinement
Based on what we saw, we made a few key adjustments. To fix the cannibalization issue, we overhauled our internal linking to make sure the short Q&A snippets always pointed prominently back to the deeper, more authoritative content. We also started consolidating some of the shorter, AI-focused pieces into bigger pillar pages, turning the Q&A format into sub-sections instead of standalone articles.
We also got serious about monitoring AI search results for our competitors. If a competitor kept getting cited in an AI overview for a query we wanted, we’d tear their content apart looking for structural differences or data points we’d missed, then update our own page to offer a better, more complete answer. This kind of competitive analysis became a routine part of our workflow.
Finally, we started to diversify our content formats. Knowing that some questions are just answered better with a video or an interactive tool, we began planning to integrate short, explanatory videos into our key AI-optimized pages. The hope is that these multimedia elements might get favored by future versions of the AI search. For a page about “how to implement a lean supply chain,” for example, we planned to embed a 90-second animated video that breaks down the steps.
The Evolving Field: Zero Traffic is Not Zero Opportunity
The LogiFlow Solutions campaign proved that while AI search absolutely changes traffic patterns, it doesn’t kill the opportunity. I think of zero traffic as ‘zero direct-click traffic for fast answers.’ Brands that learn to be the source for those fast answers can still build their name and pull in qualified leads, just through a more indirect route. Your focus has to move from just ranking high to being the most authoritative, concise, and schema-friendly answer in the entire SERP.
To have any organic visibility in the future, you have to get inside user intent and create content that answers the mail right there in the search results. That means committing to a new level of content structuring, advanced technical SEO, and constant adaptation as the AI models change. You’re collaborating with the AI to serve users better, not fighting a war against it.
What is “zero traffic” in the context of AI search?
It’s when users get their answer directly on the search results page from an AI summary, featured snippet, or “People Also Ask” box, so they never have to click through to your website. Your site might get the impression, but you don’t get the click for that specific query.
How can schema markup help mitigate the impact of zero traffic?
Schema markup is code that explicitly tells search engines what your content is about. Using specific types like FAQPage, HowTo, or a detailed Article schema makes it much more likely your content will get pulled into rich results, snippets, or AI overviews. This keeps your brand visible even if you don’t get the direct click.
Are long-tail keywords still important in an AI search environment?
Yes, they’re more important than ever. AI search is built to understand natural, conversational language. If you optimize for those very specific, question-based phrases, you have a much better shot at having your content selected as the direct answer, even if the CTR is low.
How do you measure success when users aren’t clicking through to your site?
You have to look beyond traditional clicks. Instead, track things like increases in your branded search volume (people searching for you by name), your featured snippet impressions, and how often you appear in AI overviews. You can also track indirect conversions, like when a user sees your content in an AI summary and then searches for your brand later. The “Search Appearance” report in Google Search Console becomes your best friend.
Should I put calls-to-action (CTAs) in content designed for AI search?
Yes, absolutely, but you have to be smart about it. A user who gets their answer at the top of the page won’t scroll to the bottom. You need to weave subtle, relevant CTAs into the content much earlier, right after you’ve provided a piece of value or answered a specific question. This gives them a logical next step without being obnoxious.
Are AI engines recommending your brand?
Check your score on 5 AI engines — instantly.