AI-Powered Recommendations

Revolutionary Breakthrough in AI Recommendations: RecGPT Foundation Model The recommendation systems landscape just experienced a seismic shift with the introduction of RecGPT, a groundbreaking foundation model that solves one of the industry's most persistent challenges: zero-shot generalization across domains without extensive retraining. >> The Core Innovation Traditional recommender systems hit a fundamental wall when encountering new contexts or cold-start scenarios, invariably demanding resource-intensive retraining cycles. RecGPT fundamentally departs from existing ID-based approaches by deriving item representations exclusively from textual features, enabling immediate embedding of any new item without model retraining. >> Technical Architecture Deep Dive Unified Item Tokenization: The system employs Finite Scalar Quantization (FSQ) to transform heterogeneous textual descriptions into standardized discrete tokens. This creates a domain-invariant representation that eliminates the semantic heterogeneity barriers plaguing existing systems. Each item embedding is partitioned into multiple sub-vectors, then quantized using a sophisticated pipeline involving sigmoid normalization and rounding functions with Straight-Through Estimator techniques for gradient flow. Hybrid Attention Mechanism: RecGPT implements a revolutionary dual attention system combining bidirectional attention for intra-item token processing with causal attention for sequential modeling. This enables comprehensive information sharing among tokens representing the same item while preserving temporal causality necessary for next-item prediction. Auxiliary Semantic Pathways: To counteract information loss during quantization, the framework incorporates original semantic representations alongside learned discrete token embeddings. These parallel embedding streams are combined through Layer Normalization to produce final input representations for transformer layers. Efficient Catalog-Aware Decoding: The system employs Trie-based prefix matching with beam search optimization, exploiting the sparse nature of valid item combinations within the vast token space. This transforms computationally intractable token-to-item mapping into practical real-time operations. Key Advantages: - Zero-shot transferability for immediate new item embedding - Cross-domain compatibility bridging diverse recommendation contexts   - Enhanced robustness in sparse-data environments - Real-time inference capabilities for production deployment The foundation model approach to recommendation systems has arrived, promising to transform how we think about personalization across platforms and domains. This breakthrough opens new possibilities for recommendation systems that truly understand content semantics rather than relying on brittle ID-based representations.

Surfer analyzed 289,105 URLs across ChatGPT, AI Overviews, Perplexity, and AI Mode to answer one question: Does AI recommend brands more strongly when they appear in more cited sources? Short answer: Yes. The correlation is moderately strong (0.41 Spearman). But here's where it gets interesting. ChatGPT shows the strongest correlation between brand presence and recommendation strength compared to all other AI platforms. Yet ChatGPT cites 84.7% more sources than AI Mode and 129.7% more than AI Overviews. That means getting mentioned in a high percentage of ChatGPT's cited sources is harder but more important. The second finding: Blog posts matter most. Brand mentions in blog posts correlate more strongly with recommendation strength than mentions in reviews, news articles, or any other content type. Even though blogs only make up 28.7% of cited sources, they drive the strongest recommendation impact. Third-party blogs (not your own) have the biggest effect. So what should you do? Stop trying to get mentioned everywhere. Start focusing on the blogs that AI cites repeatedly. Use tools (like Surfer's AI Tracker) to find which third-party blogs get cited most often for your target queries. Then figure out how to get mentioned in those specific posts. Reach out to authors. Pitch data they're missing. Send product samples for review. Offer expert quotes. According to Surfer's research, this is the game now. Getting your brand mentioned on frequently cited blog posts gives you the best shot at AI recommendations.

Your prospects are starting to let AI agents pick their vendor shortlists before you even know they exist. While most marketing teams optimize websites for human browsing, some buyers are already using AI agents to research, compare, and rank solutions. The change is happening now. I tested this by using ChatGPT's Agent Mode to mystery shop project management tools as if I were a marketing ops manager. The AI didn't just browse and gather information, it clicked through websites, tried to sign up for webinars, and formed clear opinions and recommendations. This goes beyond AI research tools that just analyze content. This is important for competitive positioning. Key insights from this week's newsletter: ➡︎ AI doesn't just show information, it picks favorites and tells prospects which vendor to choose ➡︎ AI weighs multiple factors to form preferences and recommend specific vendors for each situation ➡︎ You can see the exact moment when AI asks permission before handling personal information ➡︎ You'll spot competitive problems and research patterns that human browsing might miss as these tools spread ➡︎ Small differences in how you present pricing and features can swing AI recommendations in organized evaluations The mystery shopping approach gives you a practical way to walk in your customer's shoes. You'll understand what your prospects will increasingly experience and know where your competitive positioning succeeds or fails under AI review. Big thanks to David Rich (CMO at DTN) and Andy Crestodina (Co-Founder and CMO at Orbit Media Studios) for sharing their perspectives on changing buyer behavior and website experience optimization. See full breakdown in the newsletter below. To cater to different learning styles, see link in the comments for a 15-minute AI podcast version (link in the comments) of this newsletter with two AI hosts. If you found this helpful, subscribe to the newsletter (link in the comments) to get practical AI guidance every other week and feel free to share it with others.

Sharing key learnings and insights from our Real-Time (In-Session) Personalization journey at CARS24 — a capability that has transformed how we personalize the car buying experience at scale. Leveraging advanced sequence-based neural networks and real-time Kafka streaming infrastructure, we've developed a dynamic machine learning pipeline that processes more than a million user interactions daily. Our deep learning models rapidly adapt to user behaviour, delivering personalized car recommendations with sub-200ms latency. Highlights: ✅ Advanced sequence-based neural network architecture  ✅ Real-time streaming and processing of user behaviour signals with Kafka  ✅ Rapid feature engineering and inference using optimized real-time databases  ✅ High scalability for continuous model retraining and deployment Performance Impact: 📈 Across all discovery widget we achieved a highest Impression-to-View (I2V) rate and on the 'Best Matches' recommendation rail on our car detail page and buyer home page. 📈 Delivered a strong Impression-to-Booking Initiation (I2BI) conversion rate across different discovery widgets, underscoring high user relevance and engagement. Business Outcomes: 🚀 Significant uplift in user engagement  🚀 Marked reduction in user drop-offs  🚀 Enhanced personalization and superior user experience The attached flow chart outlines the architecture behind this AI-powered personalization pipeline — from real-time clickstream ingestion to ML inference and personalized recommendations. #RealTimePersonalization #AI #MachineLearning #DeepLearning #Kafka #DataScience #RecommendationEngine #TechInnovation #AI  #Personalization #pubsub #CARS24 #transformers #llm #genai

AI isn't just changing how we work—for years, it has been transforming how customers discover what they want before they even know they want it. And that's through the power of recommendation systems. Let me break it down with a simple example we've all experienced: Open any Amazon product page and you'll see: • "Frequently bought together" • "Customers who viewed this also viewed" • "Recommended based on your browsing history" These aren't just random suggestions. They're calculated revenue multipliers. When a customer lands on a product page, three things typically happen: 1. They buy the original item 2. They buy the original item PLUS recommended items 3. They skip the original item but purchase a recommended alternative Each scenario drives revenue that wouldn't exist without AI-powered recommendations. But here's what most businesses miss: This isn't just for e-commerce. Social platforms use the same concept to increase engagement. When Twitter recommends content that keeps you scrolling longer, they're not just being helpful—they're growing ad impressions and potential click-throughs. The beauty of recommendation systems? There's room for experimentation. Unlike many AI applications where precision is critical, recommendation systems can be effective even when partially accurate. If one of three suggestions resonates, that's still a win. This flexibility creates the perfect playground for companies to: 📌 Test different algorithms 📌 Experiment with recommendation placement 📌 Refine suggestion strategies without disrupting user experience Recommendation engines are a good starting point for AI experimentation—they provide a good combination of impact with minimal disruption risk. Is your business leveraging the power of "you might also like" yet?

74% of "how-to" queries now show AI Overviews instead of organic results. But that's not even the most shocking shift in search. A groundbreaking study by First Page Sage reveals something even more disruptive: ChatGPT, Gemini, Claude, and Perplexity are becoming the new gatekeepers to your customers. Welcome to the era of GEO (Generative Engine Optimization). Here's what's happening: When someone asks an AI chatbot "What's the best CRM software?" or "Who are the top marketing agencies?" - they're getting recommendations not from Google, but from AI. And the algorithm is completely different. FirstPageSage analyzed 11,128 commercial queries across the top 4 AI chatbots. The results? Mind-blowing. Here's what ACTUALLY determines AI recommendations: For ChatGPT (61.3% market share): • Authoritative list mentions: 41% • Awards & accreditations: 18% • Online reviews: 16% • Customer examples: 14% • Social sentiment: 11% For Google Gemini (13.3% market share): • Authoritative list mentions: 49% • Google website authority: 23% • Awards & accreditations: 15% • Online reviews: 13% For local businesses, it's even more fascinating: • Local business reviews: 38% • Authoritative list mentions: 29% • Online reviews: 19% This changes EVERYTHING about digital marketing. While you've been obsessing over keyword density and backlinks, your competitors might be securing spots on "top 10" lists that rank highly on Google. Why? Because when ChatGPT gets asked about your industry, it's scanning those lists to make recommendations. What this means for your business: 1. Create and rank "best of" lists that include your company (This influences 41-64% of AI recommendations!) 2. Showcase awards, certifications, and affiliations prominently (Influences 15-19% of recommendations) 3. Generate positive reviews on trusted platforms (Influences 13-31% of recommendations) 4. Publish customer case studies and usage data (Influences 13-14% of recommendations) 5. Monitor and improve social sentiment (Influences 11% of ChatGPT recommendations) The companies that master GEO now will dominate the next generation of search. Those who don't might disappear from the customer journey entirely. Our AI Agent can help you analyze your GEO potential in minutes: • Scan for list inclusion opportunities • Identify award and certification gaps • Analyze review sentiment across platforms • Generate a complete GEO strategy The future of search isn't coming. It's already here. Are you ready? Source: FirstPageSage's Generative Engine Optimization (GEO) Study (link in comments)

For years, #associations have relied on segmentation to "personalize" content. You might have: 🔹 One email for early-career professionals  🔹 One for mid-career members  🔹 Another for executives But here’s the problem: People aren’t segments. What if an early-career member is deeply invested in advanced AI applications? What if an executive wants fundamental leadership training? Segment-based personalization is outdated—and it’s leaving engagement (and revenue) on the table. The future is true 1:1 personalization, and AI is making it possible. AI-powered recommendation engines can: ✅ Understand what each member actually consumes and engages with  ✅ Deliver personalized event, course, and content recommendations  ✅ Increase engagement rates by double or triple compared to generic outreach The results? -Open rates on AI-personalized emails exceeding 100% (yes, people open them multiple times!) -Click rates that outperform industry averages -Deeper engagement and real connections between members with shared interests We’re not talking about future possibilities—we’re talking about real results happening today. Associations that embrace AI-driven personalization will see skyrocketing engagement, retention, and revenue. Those that don’t? They’ll struggle to compete. Are you leveraging AI for real personalization yet? #AI #Personalization #MemberEngagement 

ChatGPT and AI Search Matter More Than Ever I just wrapped up a call with a client who told me she found me through ChatGPT—yes, ChatGPT! While Google has long been the go-to, AI-powered search tools like OpenAI’s ChatGPT and Perplexity are now scouring the web in real time. That means if your business isn’t showing up in these AI results, you’re missing out on a whole new audience. Here’s why it matters: AI Picks from Your Content: When someone asks an AI assistant for “Find me the best headshot photographers in Jacksonville,” it pulls answers from web pages, reviews, and FAQs. If you haven’t optimized your site content or collected fresh Google Reviews, AI won’t recommend you. Conversational Discovery: People are increasingly asking AI in natural language (“Who’s a photographer near me who does corporate headshots?”). You need clear, conversational copy on your site—think Q&As, blog posts, and service descriptions that mirror real-world questions. Fresh Signals Win: AI systems reward recently updated info. Regularly add new testimonials, portfolio examples, and blog updates so AI “knows” you’re active and relevant. Bottom line: A new technology brings new opportunities, but you have to be ready to seize them. Update your website copy, encourage clients to leave reviews, and structure your content so AI can easily find and recommend you. See, AI isn't all bad! Your next client might be asking an AI—make sure it sends them your way!

I’ve spent nearly a decade building recommender systems—across both smaller orgs and big tech—developing systems from scratch and fine-tuning sophisticated recommendation engines. And along the way, I’ve learned a lot: what works, what doesn’t, and how to build something valuable without a massive budget. At the same time, I’ve been wanting to get back into writing, and now feels like the right time. My goal is to break down the complexities of AI systems into practical, no-nonsense insights that can help others on the same path. So today, I’m kicking things off with a guide on how to build a Minimum Viable Recommender System (MVR)—giving you my version of a lean, effective approach for anyone that needs results without overcomplicating things. If you want the short version, here’s the framework: 1. Set Your North Star – Align recommendations with real business goals 2. Identify Your Power Signals – Focus on the data that truly matters 3. Choose a Lean Model – Start simple and iterate as you learn 4. Build the Right Infrastructure – Balance speed, cost, and reliability 5. Validate and Iterate– Test offline first, then confirm with real users And if you want the full breakdown, click the article below! Would love to hear your thoughts—do you agree, or would you add/change anything? #recsys #recommender #ai #systems #architecture