Editor's note: Taimei AI and Mei Media are romanized names from the Chinese source, rather than independently confirmed official English brands. The 380%, fourfold, twentyfold, more-than-200% and above-90% performance figures are reported by Cai Xiaoxu and his company. The source does not provide the underlying measurements, baselines or a controlled attribution study, so this translation preserves them as interview claims.
Original · Unique Research · 2026-03-04
FOUNDER PROFILE
Infant-Formula Brand’s AI Mention Rate Surges 380%: Taimei AI Founder Cai Xiaoxu Reveals His GEO Methodology
“Compliance, authenticity, and traceability are the only way for a company to survive.”
“
“Compliance, authenticity, and traceability are the only way for a company to survive. We do not advocate pursuing short-term ‘technology dividends’; we want to build long-term ‘trust assets’ for brands.”
While most marketers are still studying how to buy traffic and acquire users, Cai Xiaoxu has led the Taimei AI team in helping an infant-formula brand increase its AI mention rate by 380%, raise its Douyin Index 4-fold, and increase its Taobao Search Index 20-fold.
This was not the work of traditional SEO, but of GEO, or Generative Engine Optimization, designed specifically for the AI era.
A veteran with more than 20 years in the marketing industry, Cai Xiaoxu previously worked at Focus Media and China Mobile 12580. In 2017, he founded Mei Media with a focus on integrated marketing. In 2023, he keenly identified a new opportunity in the AI era: users’ search habits were shifting comprehensively toward AI channels, and brands needed to establish visibility in AI-generated answers.
And so Taimei AI was born.
From Integrated Marketing to GEO: A Veteran’s AI Transformation
“My core reason for choosing to enter the AI-marketing sector was seeing two critical changes:”
Cai Xiaoxu summarizes them directly:
First, AI technology has greatly increased productivity in marketing scenarios.
Second, users’ search habits are shifting comprehensively toward AI channels, directly driving explosive growth in demand for visibility marketing.
“Every brand wants to increase exposure and establish professional recognition on AI channels. That requires professional expertise to help them place brand information precisely into AI-generated answers and capture this wave of traffic in the AI era.”
Taimei AI’s Generative Engine Optimization agent, GEOForce, was created for precisely this purpose. It is a full-stack optimization platform covering the complete closed loop from optimization strategy, intent-question identification, knowledge-graph construction, creative-model training, intelligent media distribution, and visual performance monitoring to attribution optimization.
What Is GEO, and Why Do Brands Need It?
Put simply, GEO enables brands to be seen, trusted, and chosen in AI answers and recommendations.
For example, when a user asks ChatGPT, “Which infant-formula brand is more suitable for a newborn?” or asks ERNIE Bot, “What skincare product should someone with sensitive skin use?” AI generates an answer based on training data and real-time information. GEO’s job is to ensure that brand information can be cited effectively by AI and presented to users.
But this is not merely a technical issue. As Cai Xiaoxu repeatedly emphasizes, “compliance, authenticity, and traceability” are the only way for a company to survive. Every optimization action and every performance result at Taimei AI is evidence-based, traceable, and attributable. Driven by the dual engines of “marketing + technology,” it focuses on compliant deployment and real, quantifiable outcomes, enabling brands to establish stable, sustainable exposure and professional recognition in the AI ecosystem and converting AI visibility into a brand’s long-term trust asset.
A Real Case: The Methodology Behind a 380% Increase
Cai Xiaoxu shared the case of an infant-formula brand, a representative validation of GEO’s effectiveness.
“Mothers choosing products care most about expertise and safety. By using Taimei GEO to precisely match these core needs, we planned more than 300 high-intent user questions for this formula client and implemented a systematic optimization strategy at both the brand and category levels.”
The results were impressive:
• AI mention rate increased by 380%
• Douyin Index rose 4-fold
• Taobao Search Index increased 20-fold
“In addition, our GEO solutions have delivered excellent optimization results in beauty, personal care, home appliances, and pet medicine. After optimization, brands’ mention rates increased by an average of more than 200%, while customer satisfaction and the overall renewal rate both exceeded 90%.”
Behind these figures are Taimei AI’s service capabilities. It currently serves brands in five major industries: maternity and infant products, pet medicine, personal care, small household appliances, and beauty. These include both leading brands and emerging brands deeply focused on niche markets.
The Four-Dimensional Capability Most Difficult to Replicate
In AI visibility marketing, Cai Xiaoxu believes Taimei AI’s hardest-to-replicate strength has never been a single technology, but a capability system integrating four dimensions:
1. Marketing-industry know-how—more than 20 years of accumulated hands-on marketing experience and commercialization resources.
2. Deep understanding of algorithms across platforms—mastery of the algorithmic logic and content preferences of different AI platforms.
3. Professional prompt-engineering capabilities—not simply writing prompts, but designing precise corpora around the intentions of users in a given industry.
4. A closed data-feedback loop—the complete chain from monitoring to attribution and iteration.
“
“This is an interconnected, comprehensive capability system; it cannot be achieved by piling up individual technologies.”
Which specific stage is the hardest to replicate? Cai Xiaoxu believes it is industry-specific intent stratification + the construction of a compliance-oriented knowledge graph.
“When facing the brand needs of different industries, our GEOForce first uses the logic of consumer decision-making in the industry to break down the genuine, high-intent, multilevel questions users ask across the AI-search ecosystem. It then combines industry compliance requirements to build a structured knowledge system, ensuring that AI recommendations are both precise and compliant. This requires simultaneous mastery of industry needs, compliance rules, and the logic of AI algorithms.”
The Ambition for 2026: Go Deeper and Broader, Building Long-Term Trust Assets
Taimei AI currently has two profit models:
Project-based—for large enterprises and mature brands, providing customized, full-process solutions from diagnosis through operations.
Subscription-based—for small enterprises and growing brands, enabling customers to perform multiplatform optimization independently through its Generative Engine Optimization agent, GEOForce.
In 2026, Cai Xiaoxu’s core objective is to take the GEO business deeper and broader:
“Vertically, we will thoroughly develop the major industries in which we have already implemented solutions, distill common industry problems, and form reusable industry-level GEO methodologies, compliance systems, and technical-capability architectures. Horizontally, we will rapidly expand to clients in new industries, implement brand-specific solutions on top of the industry framework, and become a leader in AI visibility marketing.”
To support this goal, Taimei AI will prioritize stronger capabilities for deep vertical-industry specialization this year, establishing a systematic self-evolving engine and a complete attribution-analysis system from content generation to business results.
“
“In the future, we hope Taimei AI can become a leader in full-stack GEO technology services, empowering brands across industries and taking them from AI visibility marketing to the conversion of commercial value—creating not merely a short-term technology dividend, but a trust asset that brands can hold over the long term.”
Selected Interview Q&A
Q1: Introduce yourself in one sentence. What are you currently working on?
A: As the founder of Taimei AI, I have worked in the marketing industry for more than 20 years and can be considered a veteran media practitioner. I am now focused primarily on GEO, or Generative Engine Optimization, specializing in providing full-stack GEO technology solutions to brand clients.
Q2: What was the core judgment behind your decision to enter the AI-marketing sector?
A: Our core reason for choosing to enter the AI-marketing sector was seeing two critical changes: on one hand, AI technology greatly increased productivity in marketing scenarios; on the other, users’ search habits were shifting comprehensively toward AI channels, directly driving explosive growth in demand for visibility marketing.
Q3: What was your background before entering AI visibility marketing?
A: I have always focused on integrated and programmatic marketing and previously worked at companies including Focus Media and China Mobile 12580. In 2017, I founded Mei Media, which primarily engaged in integrated marketing. Most of the clients we worked with were leading companies in maternity and infant products, beauty, and fast-moving consumer goods.
Q4: What prompted you to decide to build Taimei AI?
A: The decision to build Taimei GEO really came from identifying the major industry trend and market demand of the AI era—companies urgently needed a full-stack optimization solution capable of fundamentally resolving their visibility on AI channels. We had both years of accumulated marketing-industry experience and an understanding of brands and user needs, as well as validated capabilities in generative AI technology.
Q5: What is Taimei AI’s current core business model?
A: We currently position ourselves as an enterprise-grade, full-stack solution provider for GEO, or Generative Engine Optimization. With GEO as our core business, we are building an integrated GenAI service system around the entire enterprise-marketing chain.
Q6: Compared with other similar products on the market, what indispensable points of differentiation have you retained in the area of “visibility”?
A: In the core area of AI visibility, we firmly believe that “compliance, authenticity, and traceability” are the only way for a company to survive. We do not advocate pursuing short-term “technology dividends”; we want to build long-term “trust assets” for brands.
Q7: What does your typical client profile look like?
A: The typical clients we currently serve are primarily brands in five major industries: maternity and infant products, pet medicine, personal care, small household appliances, and beauty. They include both industry-leading brands and emerging brands deeply focused on niche markets.
Q8: What is the core reason clients choose you to solve their “visibility” problem?
A: Take an infant-formula brand as an example. We planned more than 300 high-intent user questions for this formula client and implemented a systematic optimization strategy at both the brand and category levels. This increased the brand’s AI mention rate by 380%; during the optimization period, its Douyin Index also rose 4-fold and its Taobao Search Index increased 20-fold.
Q9: In AI visibility marketing, what capability do you believe is the hardest to replicate?
A: In AI visibility marketing, or GEO, I believe our hardest-to-replicate strength has never been a single technical capability. It is the four-dimensional integration of marketing-industry know-how, a deep understanding of algorithms across platforms, professional prompt-engineering capabilities, and a closed data-feedback loop.
Q10: Can you give an example of a specific stage or client scenario that “others would find very difficult to copy”?
A: I believe the hardest stage of Taimei GEO to replicate is industry-specific intent stratification + the construction of a compliance-oriented knowledge graph. This requires simultaneous mastery of industry needs, compliance rules, and the logic of AI algorithms; such comprehensive deployment capabilities are very difficult to copy.
Q11: What are the main ways Taimei AI’s GEO business currently generates revenue?
A: Our profit model is divided mainly into two types: one is project-based GEO service delivery for large enterprises and mature brands; the other is GEOForce for small enterprises and growing brands.
Q12: What internal metrics do you use to judge whether “this monetization path is healthy”?
A: We have an internal dual-track validation system. For the project-based model, the core measures are high-value client retention and renewal rates. For the subscription-based model, in addition to customer renewal rates, we also monitor GEOForce activity.
Q13: Across the full chain from “prompt engineering” to “commercial conversion,” which stage has the highest loss rate?
A: Across the full GEO-deployment chain, the stage with the highest loss rate is actually the matching problem between “prompt engineering” and “effective exposure.” If prompt engineering does not incorporate actual user intent in the industry, and the solution lacks the ability to track the algorithmic logic of different AI platforms, then the content will be difficult for AI to output effectively.
Q14: What is the measurement framework you most commonly use for “visibility conversion efficiency”?
A: Our metrics for measuring GEO visibility conversion efficiency include multiple dimensions: coverage of core questions, the AI citation rate of media publications, AI’s mention rate for brands or products, and the accuracy of AI-recommended content.
Q15: AI allows marketing content to be generated without limit, but platforms’ visibility algorithms are also evolving rapidly. How do you remain effective?
A: To address this, we developed an AI-visibility monitoring and diagnostic system equipped with a multidimensional OLAP analysis engine. This effectively builds a closed data-feedback loop across the entire GEO chain, forming a deployment loop of “intent decomposition → content generation → performance monitoring → strategy iteration.”
Q16: In clients’ actual processes, which stages does AI prompt engineering make “faster,” and which instead require “human reinforcement”?
A: AI prompt engineering has indeed produced a qualitative improvement in efficiency for many repetitive, scalable marketing stages. But when dealing with different media’s publication rules and content-compliance reviews, AI currently finds it difficult to achieve 100% precise control, so these stages still need human reinforcement.
Q17: What commercialization breakthrough do you value most in 2026?
A: Our most important commercialization breakthrough this year is to take the GEO business deeper and broader: vertically, to thoroughly develop the major industries in which we have already implemented solutions; and horizontally, to rapidly expand to clients in new industries and become a leader in AI visibility marketing.
Q18: Which capability will you prioritize strengthening to achieve this breakthrough?
A: This year, we will prioritize strengthening our capacity for deep vertical-industry specialization, continuously improving knowledge graphs and prompt-engineering asset libraries for industry subsegments and building media-resource networks, content industrialization, and a self-evolving engine for the GEO system.
Q19: Predict in one sentence: what kind of company will Taimei AI become by the end of 2026?
A: In the future, I hope Taimei AI can become a leader in full-stack GEO technology services, empowering brands across industries and taking them from AI visibility marketing to the conversion of commercial value. We are steadily moving in that direction.
Material sourced from a Unique Awards guest interview.