The debate over how digital marketers and SEO professionals should measure visibility across generative artificial intelligence platforms intensified following a recent declaration by Siege Media CEO Ross Hudgens. In a viral LinkedIn post, Hudgens argued that Perplexity’s shrinking market share invalidates its inclusion alongside foundational AI models like ChatGPT, Gemini, Claude, and Google’s ecosystem search products. Hudgens urged his peers to completely remove Perplexity from their large language model (LLM) tracking software, warning that treating it as an equal peer skews performance analytics and misallocates strategic priorities.
However, a closer examination of industry data, historical search engine evolution, and multi-tier market dynamics reveals a more nuanced reality. While Perplexity has unquestionably lost ground in raw referral traffic and general web share over the course of 2026, completely discarding niche or emerging platforms from tracking metrics risks repeating historic missteps in search engine optimization.
Understanding the Evolution of Search Fragmentation
The current state of AI search mirrors historical shifts in the digital landscape. Decades ago, during the formative years of modern search engine optimization, industry professionals regularly debated whether to report rankings across dozens of fragmented engines or consolidate tracking around established legacy players. In March 2002, amid the aftermath of the dot-com crash, early search visibility reports often tracked as many as 15 distinct platforms. Conferences of that era frequently featured debates advocating for the simplification of reporting by focusing strictly on the dominant incumbent engines of the day—such as Yahoo, Excite, Lycos, AltaVista, and Ask Jeeves.
Crucially, those early consolidation proposals frequently omitted Google, which at the time was an aggressive, fast-growing challenger rather than the entrenched monopoly it eventually became. Dismissing rising platforms prematurely because they lacked immediate dominance proved to be a costly strategic error for agencies that failed to adapt.
Applying this historical lens to the current generative AI boom suggests that the market is not settling into a simple, permanent duopoly or oligopoly. Instead, search behavior is splintering into distinct functional tiers, where raw traffic volume tells only part of the story. The primary challenge for modern SEO professionals is not merely identifying which platforms command the most eyeballs today, but recognizing which tools are scaling fast enough—or serving specialized enterprise niches deeply enough—to alter how consumers and businesses discover information.
Divergent Data: Referral Share Versus Web Traffic
Evaluating Hudgens’ thesis requires separating different metrics of success: web traffic volume versus actual referral share and strategic utility. Publicly available internet telemetry illustrates why industry analysts hold conflicting views on Perplexity’s market standing.
Data from StatCounter tracking worldwide AI chatbot referral shares throughout mid-2026 highlights a significant contraction for Perplexity. In June 2026, Perplexity commanded approximately 7.91% of AI chatbot referral share, putting it in a virtual dead heat with Google Gemini, which sat at 7.94%. However, by August 2026, Perplexity’s referral share tumbled to 4.31%, while Gemini surged upward to capture 10.9% of referrals.
Concurrently, comprehensive web traffic assessments from Similarweb present a broader picture of absolute audience reach among major AI assistants. In May 2026 metrics evaluating the top generative platforms, OpenAI’s ChatGPT maintained a commanding lead with roughly 53.9% of worldwide web visits among major assistants, followed by Google Gemini at 27.9%, Anthropic’s Claude at 9.2%, DeepSeek at 4.1%, xAI’s Grok at 2.4%, and both Perplexity and Microsoft Copilot trailing at 1.3% each.
These divergent figures validate the core of Hudgens’ technical concern: if an automated SEO reporting dashboard assigns equal weight to every integrated LLM, an anomalous performance spike or drop on a low-traffic platform like Perplexity can disproportionately distort an enterprise’s overall visibility score. When a brand’s aggregate score relies on an unweighted average, a minor player with low actual user volume can artificially inflate or deflate the apparent health of an SEO campaign.
The Emergence of a Three-Tier Market Structure
Rather than a unified ecosystem where every AI tool functions interchangeably, the generative search landscape has rapidly stratified into a three-tier market structure characterized by distinct distribution channels and audience intents.
Tier One consists of foundational general-purpose models with massive consumer and developer scale: ChatGPT, Gemini, and Claude. OpenAI reported reaching over one billion active users globally across its product ecosystem by mid-2026, cementing its direct consumer dominance. Google integrated Gemini deeply across its massive hardware and software footprint, including Android and core search properties, pushing the Gemini app past one billion monthly users by late summer 2026. Meanwhile, Anthropic carved out a powerful position in professional and developer workflows, reporting annualized revenue climbing to $65 billion by July 2026, driven by aggressive enterprise adoption and high-value corporate partnerships.
Tier Two comprises ambient AI features embedded directly into legacy search and productivity ecosystems. Google AI Overviews and AI Mode represent a fundamentally different category of discovery than standalone conversational chat interfaces. By mid-2026, Google reported that AI Overviews reached over 2.5 billion users monthly, while AI Mode surpassed one billion monthly users, appearing in roughly 43% of U.S. search queries. Similarly, Microsoft Copilot leverages deep integration within the Microsoft 365 enterprise suite, serving hundreds of millions of corporate and consumer users. Treating these ecosystem-level features as mere chatbot peers fundamentally underestimates their influence on traditional search journeys.
Tier Three encompasses emerging, specialized, or declining platforms, including Perplexity, Grok, and DeepSeek. While these applications command smaller absolute shares of total web traffic, they often retain outsized importance within specific demographics, research-heavy verticals, or unique monetization models. Perplexity, for example, continues to pursue a distinct subscription and enterprise-focused strategy without traditional display advertising, maintaining active industry partnerships—such as integrating Similarweb market intelligence into Perplexity Computer—that signal ongoing commercial viability despite smaller consumer traffic numbers.
Enterprise Adoption and the Limits of Traffic-Only Metrics
A critical flaw in relying solely on consumer web traffic to determine tracking priorities is that it overlooks specialized B2B and enterprise utility. Anthropic’s Claude exemplifies this phenomenon. While its consumer-facing web traffic share trails behind ChatGPT and Gemini, its deep penetration into major professional services firms and global enterprises changes the equation for B2B marketers.
Major corporate deployments highlight this trend. Professional services giants such as PwC expanded partnerships to deploy Claude Code and Cowork across tens of thousands of global professionals, while Tata Consultancy Services (TCS) and Cognizant initiated large-scale enterprise rollouts and certification programs for tens of thousands of employees. Furthermore, Anthropic reported thousands of enterprise customers spending upwards of $1 million annually.
For a B2B SaaS provider, a cybersecurity vendor, or an enterprise-focused consultancy, the platform driving consumer web traffic may be entirely irrelevant compared to the model utilized inside corporate developer environments and boardroom workflows. Consequently, B2C retailers, software companies, and industrial manufacturers require vastly different LLM tracking portfolios tailored to their specific target audiences.
Implications for SEO Measurement and Visibility Scoring
The debate over Perplexity underscores a broader methodological challenge facing modern digital marketing: the danger of relying on aggregate visibility scores.
Consider a hypothetical brand achieving a 40% citation rate on ChatGPT, 35% on Gemini, 30% on Claude, and 90% on Perplexity. A simplistic, unweighted average yields a misleading visibility score of nearly 49%. If Perplexity accounts for only a fraction of a percent of actual customer acquisition or referral traffic for that specific business, its exceptionally high citation rate artificially skews executive reporting.
To construct a reliable measurement framework, analytics professionals must connect three distinct datasets:
- Audience Exposure: Monitoring underlying platform usage, web visits, and demographic distribution to understand where target audiences actually spend time.
- Platform Visibility: Tracking brand mentions, citations, linked URLs, and the specific prompt parameters that generate them across individual models.
- Business Impact: Correlating AI-driven referral traffic and direct citations down-funnel to engagement metrics, qualified leads, and actual revenue conversions.
Revisiting the Prescription for Marketers
The emerging consensus among data-driven search marketing experts suggests that completely deleting Perplexity or other niche tools from tracking suites is an overcorrection. While Hudgens correctly identified the distortion caused by unweighted multi-model averages, the correct remedy is de-weighting, not deletion.
If Perplexity or any emerging search assistant represents a negligible slice of a brand’s verified referral ecosystem, it should account for a proportionally tiny fraction of an aggregate visibility metric. However, if a specific niche or industry vertical experiences disproportionate discovery through alternative engines, completely removing those platforms risks blinding marketing teams to early market shifts and competitive threats.
Ultimately, the lesson of modern AI search mirrors the lessons of the early 2000s search engine boom: market leaders evolve rapidly, distribution models shift without warning, and dismissing challengers prematurely can leave brands unprepared for the next major transformation in digital discovery. Track the leaders, weight metrics according to real-world traffic and business impact, but maintain a watchful eye on the innovators operating at the edges of the ecosystem.
