In the fast-evolving landscape of artificial intelligence, a provocative debate has emerged regarding how marketers and SEO professionals should measure their brand’s digital footprint. The controversy was ignited by a recent LinkedIn post from Ross Hudgens, CEO of the content marketing agency Siege Media, who argued that Perplexity’s waning market share renders it an unnecessary variable in modern LLM (Large Language Model) tracking.
Hudgens’ directive was blunt: "Everyone should remove Perplexity from their LLM trackers today." His argument rests on the premise that including Perplexity alongside titans like ChatGPT, Gemini, and Claude provides a distorted, inaccurate view of where true audience visibility occurs. However, this assertion has sparked a deeper industry conversation about whether simplifying our metrics—or ignoring emerging players—is a tactical error that mirrors the mistakes of the early 2000s search era.
The Main Facts: The Case for Consolidation
At the heart of this debate is a shift in how users interact with AI. As the novelty of generative AI wears off, usage is coalescing around platforms that offer deep ecosystem integration or specialized utility.
Hudgens’ core concern is data hygiene. If a marketing platform weights Perplexity equally with ChatGPT, a single anomalous week of high performance on the former could lead a CMO to believe their brand strategy is succeeding, even if the primary channel (ChatGPT) is failing. In a world where data-driven decisions are paramount, "noisy" metrics from platforms with shrinking reach can mask reality.
However, the counter-argument is that market share is a snapshot, not a crystal ball. While Perplexity has seen a decline in its referral share, it remains a distinct tool for a specific type of information-seeking behavior. The fundamental question for the industry is no longer "which AI is biggest?" but "which AI platforms are essential for a specific business model?"
Chronology of a Shifting Landscape
To understand the current tension, we must look at the trajectory of the AI search market throughout 2026.
- Early 2026: Perplexity maintained a competitive edge, positioning itself as the "answer engine" for power users and researchers. At this stage, it was frequently included in the "Big Four" of LLM tracking.
- June 2026: The turning point. StatCounter data indicated that Perplexity accounted for 7.91% of AI chatbot referral share, nearly identical to Google’s Gemini (7.94%).
- August 2026: A sharp divergence occurred. Perplexity’s referral share plummeted to 4.31%, while Gemini surged to 10.9%. This decline, captured by StatCounter, provided the fuel for Hudgens’ critique.
- September 2026: The debate moved from internal agency discussions to the public square, with leaders like Hudgens calling for the systematic removal of "fringe" models from reporting dashboards.
Supporting Data: Understanding the "Big Three" vs. The Rest
While Hudgens suggests that Perplexity is losing relevance, the broader data suggests a more complex story of an "emerging oligopoly." Similarweb’s mid-2026 data provides the most granular look at this shift:
- ChatGPT: Despite a decline in overall traffic share—from 76.4% in 2025 to 52.7% in May 2026—it remains the dominant force, with OpenAI reporting over one billion active users across its product suite.
- Gemini: The primary beneficiary of the recent shifts. By leveraging Google’s massive distribution network across Android and Search, Gemini has climbed from roughly 9% to 27.3% of traffic share.
- Claude: Anthropic’s model has solidified its position in the B2B and enterprise sectors. With 100,000+ customers on Amazon Bedrock and reports of $65 billion in annualized revenue as of mid-2026, Claude has carved out a "professional" niche that consumer-facing metrics often fail to capture.
The data indicates that we are not moving toward a single winner, but rather a trifecta of distribution advantages: ChatGPT for general consumer adoption, Gemini for ecosystem integration, and Claude for professional and enterprise workflows.
The "Google" Problem: A Separate Tier of Visibility
One of the most critical points in this debate is the status of Google’s own AI features. AI Overviews and "AI Mode" are not simply another LLM to be tracked—they are a fundamental shift in how the world’s most dominant search engine operates.
By May 2026, AI Overviews were appearing in 43% of U.S. Google searches, up from 15% just a year prior. With 2.5 billion users interacting with these AI-enhanced results monthly, treating Google’s AI as a standard LLM is a categorical error. For SEOs, these features represent an "AI layer" that must be tracked as a distinct category, separate from the chat-based models like ChatGPT or Claude.
Implications: The Three-Tier Measurement Framework
Rather than deleting platforms, the industry should move toward a tiered, weighted measurement framework. This approach acknowledges the reality of current traffic while remaining vigilant against future disruption.
Tier 1: The Anchors (Scale & Strategic Significance)
This tier includes ChatGPT, Gemini, and Claude. These models have the highest volume of general and professional traffic. Marketers should track these individually rather than as a singular "AI" metric.
Tier 2: The Ecosystem Layers (Search-Discovery)
This includes Google AI Overviews, AI Mode, and Microsoft Copilot. These are not just chatbots; they are the new front end of the internet. Their measurement should be integrated into existing search performance metrics, acknowledging that they are altering the traditional search journey.
Tier 3: The Specialized & Emerging (Monitoring)
This is where Perplexity, Grok, and DeepSeek belong. These platforms should not be given equal weight in a dashboard, but they should absolutely remain in the tracking portfolio.
The "2002 Flashback" and the Danger of Hubris
The skepticism toward simplifying AI tracking is rooted in historical precedent. In 2002, when the dot-com bubble burst, many SEO professionals proposed focusing only on the "Big Five" search engines of that era (Yahoo, Excite, Lycos, AltaVista, and Ask Jeeves). Those who followed this advice missed the rapid rise of a small, upstart company called Google.
The lesson from 2002 is that "small" is not a permanent state. Perplexity, despite its recent share loss, continues to innovate in the ad-free, subscription-based, and intelligence-heavy space. Similarweb’s ongoing partnership with the company to provide market intelligence suggests that there is still significant strategic value in the platform.
If an SEO team ignores a platform because it represents only 1% of their current traffic, they risk being blind to the "slow-burn" growth that could eventually redefine the market.
Final Verdict: Don’t Delete, Just Re-Weight
Ross Hudgens is correct that our current measurement models are flawed. Using a simple, unweighted average to track AI visibility creates a "false sense of precision" that is dangerous for strategic planning.
However, the solution is not erasure. It is contextual weighting.
If Perplexity accounts for only a fraction of your traffic, treat it as such. Give it 1% of the weight in your reporting, not 25%. If your business is in a niche where Perplexity users are your primary audience—such as high-end research or academic consulting—then its weight should be significantly higher.
The goal of modern marketing intelligence is to connect three distinct data points:
- Audience Exposure: Where are your potential customers spending their time?
- Visibility: How often does your brand appear in the answers generated by these platforms?
- Business Impact: Does that visibility translate into leads, sales, or brand equity?
By moving away from "aggregate LLM scores" and toward a weighted, platform-specific approach, marketers can maintain the agility to track the leaders of today while keeping a watchful eye on the disruptors of tomorrow. In the high-stakes game of AI search, the platform you dismiss today might be the one you regret ignoring when the market shifts again.
