If your organic traffic dropped 20-40% over the past year and you can’t figure out why, AI search is likely the culprit. Google AI Overviews, ChatGPT, Perplexity, and Gemini now answer queries directly, which means fewer clicks reach your website even when your rankings haven’t moved.
This isn’t a temporary blip. It’s a structural shift in how people find information. The good news? You don’t have to sit and watch your traffic disappear. There’s a playbook for this, and the brands executing it are actually growing faster than before.
At upGrowth, we’ve helped clients like Fi.Money and Vance not just survive but dominate AI search visibility.Fi.Money became the top cited authority for smart deposit queries in Google AI Overviews, gaining 200K additional clicks and 7M new impressions. That didn’t happen by accident.
Here’s what’s actually going on and what you can do about it.
In This Article
Share On:
Why is AI taking my website traffic?
The core problem is zero-click search at scale. When someone asks “how to reduce customer acquisition cost,” Google AI Overviews now synthesizes an answer from multiple sources and displays it right in the search results. The user gets what they need without clicking through.
ChatGPT and Perplexity take this further. They don’t just summarize search results; they generate complete answers, sometimes citing sources and sometimes not. Your carefully crafted blog post might be the training data behind an AI answer, but you get zero credit and zero traffic.
Here’s the data that should worry you: in 2025, zero-click searches accounted for roughly 60% of all Google queries. For informational queries (the bread and butter of content marketing), that number was even higher.
But here’s what most businesses miss. AI engines still need sources. They still cite content. The brands that get cited capture a new kind of traffic: AI-referred traffic. It’s smaller in volume but significantly higher in intent.
The traffic you’re losing vs. the traffic you should chase
Not all lost traffic is equal. Most of the traffic AI “stole” was low-intent informational traffic that rarely converted anyway. Someone searching “what is content marketing” was never going to buy your services. They wanted a definition, and now AI gives it to them instantly.
The traffic you should actually care about falls into two buckets.
Decision-stage queries where buyers evaluate options. “Best SEO agency for fintech startups” or “fractional CMO vs full-time CMO cost.” AI answers these too, but it cites specific brands when answering. If you’re not the brand being cited, your competitor is.
Problem-aware queries where someone describes a pain point. “My website traffic is declining what to do” or “Google Ads CPL too high how to fix.” These searchers want actionable solutions, not definitions. AI answers them with frameworks and recommendations, often linking to the source.
upGrowth’s approach with Vance exemplifies this. Instead of chasing vanity traffic, we optimized for geo-targeted payment queries where AI Overviews dominate. The result was 70% traffic growth because we targeted the queries where AI citations actually send clicks.
Step 1: Audit what AI engines say about you right now
Before you fix anything, you need to know where you stand. Open ChatGPT, Perplexity, Google AI Overviews, and Gemini. Search for 10-15 queries your ideal customers would ask.
Document three things for each query. First, does AI mention your brand at all? Second, which competitors get mentioned or cited? Third, what sources does AI pull from?
If you’re invisible across all four platforms, that’s your baseline. If competitors show up and you don’t, that tells you exactly what content gaps to prioritize.
This audit typically reveals one of three situations. You have no content that AI can extract (content gap). You have content but it’s not structured for AI extraction (format problem). Or you have content, it’s well-structured, but your domain authority isn’t strong enough for AI to trust it (authority problem).
Each situation requires a different fix. Most businesses have all three problems to varying degrees.
Step 2: Restructure content for AI extraction
AI engines don’t read content the way humans do. They scan for extractable statements, which are complete, quotable sentences that directly answer a query. If your content buries the answer in paragraph three after a long introduction, AI will skip you and cite someone who puts the answer upfront.
The fix is what we call BLUF structure: Bottom Line Up Front. Every section of every page should lead with the direct answer in the first two sentences. Supporting evidence, nuance, and context come after.
Here’s a practical example. Say you’re targeting “how to reduce Google Ads CPL.” Don’t start with a paragraph about the importance of cost optimization. Start with: “To reduce Google Ads CPL, audit your search term report weekly, pause keywords with CPL 2x above your target, and shift budget to exact match variants of your top converters.”
That sentence is extractable. AI can quote it directly and attribute it to you. The rest of your content adds depth, but that opening sentence is what wins the citation.
Every H2 section on your site should pass this test: if an AI engine pulled only this section and showed it to a user, would they get a complete, useful answer?
Step 3: Build content that AI must cite (not just reference)
Generic content gets summarized but not cited. Original content gets cited because AI can’t generate it independently.
What counts as original? Proprietary data from your business. Case studies with specific metrics. Frameworks you’ve developed. Survey results you’ve published. Benchmarks from your client base.
When upGrowth writes about fintech marketing, we don’t say “digital marketing is important for fintech companies.” We say “Lendingkart achieved 5.7x lead volume increase while reducing CPL by 30% through a combination of granular audience targeting and landing page optimization.” AI can’t make that up. It has to cite us.
Build a library of these proof points. For every service you offer, you should have at least three data-backed claims that only you can make. These become your citation anchors.
Step 4: Optimize for AI crawlability
This is the technical layer most businesses completely ignore. AI bots (OAI-SearchBot for ChatGPT, PerplexityBot, ClaudeBot, Google-Extended for Gemini) need explicit permission to crawl your site.
Check your robots.txt file right now. If it blocks any of these bots, you’re invisible to those AI platforms by design. Many enterprise sites accidentally block AI crawlers because their security teams added blanket bot restrictions.
Beyond access, your site structure matters. Clean URLs, proper heading hierarchy, schema markup (especially Article, FAQPage, and HowTo schemas), and fast load times all signal to AI crawlers that your content is trustworthy and extractable.
One client came to us after their traffic dropped 35% in six months. The first thing we found? Their CDN was blocking all AI crawlers. A single robots.txt fix started recovering their visibility within weeks.
Step 5: Create an AI citation measurement system
You can’t improve what you don’t measure. Traditional SEO metrics (rankings, organic sessions, keyword positions) don’t capture AI visibility. You need new metrics.
Track AI-referred traffic using UTM parameters. Configure your analytics to capture visits from chatgpt.com, perplexity.ai, gemini.google.com, and other AI referrers. This tells you how much traffic AI sends today.
Run monthly citation audits. Search your top 20 target queries across ChatGPT, Perplexity, and Google AI Overviews. Record whether you’re mentioned, cited, or absent. Track this over time to see if your optimization efforts are working.
Measure citation share: of all the sources AI cites for your target queries, what percentage are yours vs competitors? This is the new “market share” metric for the AI era.
upGrowth built this measurement infrastructure for Fi.Money, which is how we documented the 200K click increase and 7M impression growth. Without measurement, you’re optimizing blind.
What NOT to do: common panic reactions
Don’t block AI crawlers: Some businesses think “if AI can’t crawl my content, users will have to visit my site.” Wrong. AI will just cite your competitor instead, and you lose both AI visibility and the traditional traffic.
Don’t stuff your content with brand mentions hoping AI will repeat them: AI engines are sophisticated enough to distinguish between genuine authority signals and keyword stuffing.
Don’t abandon content marketing entirely: Yes, informational traffic is declining. But content is still the primary mechanism through which AI engines discover and evaluate your brand. The solution is better content, not less content.
Don’t chase every AI platform simultaneously: Start with the one that matters most for your audience. For B2B, that’s typically Google AI Overviews and Perplexity. For consumer brands, ChatGPT and Google AI Overviews dominate.
The bigger picture: AI search is a distribution channel, not a threat
Reframe how you think about this. AI search isn’t destroying your business model. It’s creating a new distribution channel that rewards authoritative, original, well-structured content.
The brands that win in 2026 and beyond aren’t the ones with the most content. They’re the ones AI engines trust enough to cite. Building that trust requires the same things that have always mattered: genuine expertise, original insights, proven results, and clear communication.
The difference is that now, those signals need to be machine-readable, not just human-readable. That’s the entire GEO discipline in one sentence.
Conclusion
AI search isn’t killing website traffic universally. It’s redistributing traffic from low-intent informational queries to high-intent AI-referred visits. The brands seeing traffic declines are those still optimized for 2023-era search behavior. The brands growing are those that adapted their content strategy for AI citation.
The structural shift is zero-click search at scale, where 60% of Google queries in 2025 didn’t result in clicks. For informational queries, that number was even higher. But AI engines still need sources. The traffic you should chase isn’t vanity informational traffic. It’s decision-stage queries where AI cites specific brands and problem-aware queries where AI links to actionable solutions.
The five-step recovery playbook starts with auditing what AI engines currently say about you, then restructuring content for AI extraction using BLUF principles, building citation-worthy original content with proprietary data, optimizing technical crawlability for AI bots, and creating measurement systems to track citation share over time.
upGrowth’s work with Fi.Money and Vance demonstrates what’s possible when AI visibility becomes a strategic priority rather than a reactive concern. Fi.Money’s 200K click increase and 7M impression growth came from systematic GEO optimization, not hoping AI would randomly discover their content.
Recover your AI search visibility
If your traffic declined 20-40% and AI displacement is the likely cause, the solution isn’t hoping traffic returns. It’s adapting your content strategy for AI citation through Generative Engine Optimization.
upGrowth pioneered GEO as a standalone service because we saw this shift coming in 2024 and built the methodology early. Our clients who invested in AI visibility while competitors waited are now capturing citation share that compounds into sustainable competitive advantage.
Contact us to discuss your AI visibility strategy. We’ll audit where you currently stand across ChatGPT, Perplexity, Google AI Overviews, and Gemini, then show you the specific optimizations that will recover and grow your traffic.
Frequently asked questions
1. How much traffic are websites losing to AI search in 2026?
Most content-heavy sites report 20-40% organic traffic decline year-over-year, primarily on informational queries. However, sites optimized for AI citations often see net positive growth because AI-referred traffic converts at 2-3x the rate of generic organic traffic.
2. Can I get my old traffic back?
The informational traffic that AI now handles directly isn’t coming back. But you can replace it with higher-quality AI-referred traffic by optimizing for citations. The total volume may be lower, but the conversion value is significantly higher.
3. How long does it take to see results from GEO optimization?
Most clients see initial AI citation improvements within 6-8 weeks of implementing structural content changes. Meaningful traffic recovery from AI channels typically takes 3-4 months, similar to traditional SEO timelines.
4. Does this only affect informational content?
Informational content is hit hardest, but transactional and comparison queries are increasingly influenced too. Google AI Overviews now appear for roughly 30% of transactional queries, making GEO optimization critical for all content types.
5. Should I hire a GEO specialist or can my SEO team handle this?
GEO requires skills that overlap with SEO but go further: AI prompt engineering, citation audit methodology, extractable content architecture, and multi-platform optimization. Most SEO teams can learn these skills, but the learning curve is steep. A specialist accelerates results significantly.
Amol has helped catalyse business growth with his strategic & data-driven methodologies. With a decade of experience in the field of marketing, he has donned multiple hats, from channel optimization, data analytics and creative brand positioning to growth engineering and sales.
Growth Strategy and Planning
