Published Jainuary 23, 2025
1
Natural Conversations
The essence of true personalisation lies in natural, human-like interactions. Every shopping trip with my wife Bhavana demonstrates this perfectly. Through fluid conversations with sellers, she articulates preferences, combines elements from different items, and guides the discovery process until finding exactly what she wants. “This colour, that design, and that fabric” – simple instructions that sellers instantly understand and act upon, creating a genuine N=1 experience.
This natural customisation extends across all real-world shopping experiences. Whether choosing a saree, selecting luggage, or ordering at restaurants (where I want my sev puri without tomatoes, and Bhavana prefers hers extra spicy), human sellers grasp our individual preferences through simple conversation. They understand context, remember our choices, and adapt their recommendations accordingly. Most importantly, they put our interests first, knowing that customer satisfaction drives long-term relationships.
Yet this natural, conversational personalisation remains frustratingly absent in our digital interactions with brands. I cannot simply tell Amazon, “Alert me when any of my favourite ten thriller authors releases a new book, especially legal thrillers like Grisham and Turow. Also, keep me updated on good new books about AI and business.” While recommendation algorithms work reasonably well, they lack the nuanced understanding that comes from direct dialogue, and can be easily derailed by casual browsing.
Similarly, Netflix doesn’t allow me to say, “I have two hours free and want a movie mixing comedy and action, with a touch of romance” – a request any friend could easily handle. Instead, we’re confined to rigid category navigation and limited search functionality that often misses the context and nuance of our preferences.
This isn’t about deploying better chatbots to search product catalogs. It’s about creating intelligent agents that truly understand us as individuals and can engage in meaningful dialogue. While we can refine and shape outputs from AI models like ChatGPT and Claude, we cannot easily do the same with search engines and recommendation systems.
Enter MyTwin – an AI-powered digital mirror of ourselves which resides inside the brand backend. It learns, anticipates, and even surprises us with its understanding. Unlike traditional recommendation engines, MyTwin prioritises our interests above all else. Through natural conversations, it builds a deep understanding of our preferences, acting as our personal agent in interactions with the brand. Each interaction refines its knowledge, creating an ever-more-accurate reflection of our desires and interests.
Together with NeoMails, MyTwin forms the second pillar of the NeoAdtech future – where B2C products are offered to businesses for free and monetised through precisely targeted PII-based advertising. This combination has the potential to redirect hundreds of billions in AdWaste toward more effective marketing, helping brands transform from being profitless to achieving profipoly status. More importantly, it could lay the foundation for creating the next trillion-dollar enterprise by fundamentally reimagining how brands and customers interact in the digital age.
The implications are profound: MyTwin could finally bridge the gap between the natural personalisation we experience in human interactions and the rigid, algorithmic approaches that dominate digital commerce today. It’s not just about better recommendations – it’s about creating truly personal relationships between brands and customers, mediated by AI that puts individual preferences first.
2
Beyond Segments
Marketers treat customers as part of a segment or cohort. A single customer may be part of multiple such groupings. For example, I might be categorised as a premium credit card holder, frequent air traveller, business book buyer, thriller fiction reader, tech gadget early adopter, and news junkie. These segments help brands target offers and customise communications, but they miss the nuanced reality of individual preferences and behaviours.
The limitations of segment-based marketing become apparent in daily interactions. Amazon might recommend business books I’ve already read or thrillers from authors I don’t enjoy. Netflix suggests movies similar to ones I watched but didn’t particularly like. Food delivery apps keep pushing cuisines I rarely order. The fundamental problem is that these systems treat me as a collection of segments rather than as a unique individual with specific preferences, dislikes, and context.
Even sophisticated AI-powered recommendation engines fall short because they rely primarily on observed behaviour – what I clicked, watched, or purchased. But past actions don’t always reflect true preferences. A one-time purchase for a gift shows up as an interest. A movie watched with family becomes part of my taste profile. A random click while browsing influences future recommendations. Without the ability to directly communicate preferences and context which can create a truer representation and understanding of me, these systems accumulate biases and misunderstandings over time.
The solution requires moving beyond segments to true N=1 personalisation where each customer is understood as a unique individual. This is where MyTwin becomes transformative – it creates a personalised AI agent that learns directly from conversations with the customer, understands context and preferences explicitly rather than just inferring them from behaviour, and acts as an intelligent interface between individual desires and brand offerings.
Most importantly, MyTwin preserves and builds upon this understanding over time rather than being led astray by occasional deviations in browsing or purchase patterns. It’s like having a personal shopper who knows you deeply, remembers your preferences, and can engage in natural dialogue to refine recommendations and suggestions.
This shift from segment-based to truly personal AI agents represents the next evolution in marketing – one that finally delivers on the promise of genuine N=1 engagement at scale.
3
N=1
Marketing automation has evolved from broad segments (N=1000s) to micro-segments (N=100s), but fundamentally remains trapped in a segmentation mindset. Companies invest heavily in sophisticated marketing automation platforms that slice audiences into increasingly narrow segments, yet struggle with low retention and limited revenue per customer. Despite employing advanced techniques like collaborative filtering and psychographic segmentation, they fail to build lasting customer relationships. A streaming service might correctly identify a customer’s favourite genres but miss understanding how those viewing preferences change based on who they are watching with. A travel platform may categorise someone as a luxury traveller but fail to recognise that their splurges are only for annual family trips, while they prefer budget options when traveling solo.
True N=1 personalisation is fundamentally different. It’s not about creating smaller segments or more precise cohorts – it’s about understanding and serving each customer as a unique individual with specific contexts, preferences, and goals. Just as a personal shopper builds deep knowledge of an individual client’s tastes, or a financial advisor understands their client’s unique circumstances and aspirations, N=1 personalisation requires moving beyond observed behaviours to genuine understanding.
The challenge isn’t just technical – it’s conceptual. Traditional martech platforms focus on operational efficiency and transaction-focused campaigns: how many messages to send, when to send them, which products to recommend based on past behaviour. But true N=1 personalisation requires blending automation with authentic human-like understanding – not just streamlining operations but improving the quality and relevance of every interaction and thinking of it as a journey. It requires understanding not just what customers have done, but what they want to achieve.
This is where MyTwin becomes transformative. Instead of trying to fit customers into ever-smaller segments, it creates a personalised AI agent that learns directly from conversations with the customer. Like a GPS system that transforms static map data into personalised navigation guidance by understanding the specific destination and preferences, MyTwin transforms a brand’s products, services, and content into enriching experiences that align with each customer’s unique journey.
Most importantly, MyTwin preserves and builds upon this understanding over time rather than being led astray by occasional deviations in browsing or purchase patterns. It enables what I call “extreme personalisation” – where every interaction is genuinely tailored to the individual’s context and goals, not just their segment characteristics.
The shift from segmentation to true N=1 personalisation represents more than an incremental improvement in marketing technology. It’s a fundamental reimagining of how brands build customer relationships – one that finally delivers on the promise of treating each customer as a unique individual rather than a collection of segment attributes.
PS: I want to thank Prasad Ram, co-founder of Gooru Learning, for having patiently explained to me the true meaning of N=1 personalisation, and how it differs from what we think of us personalisation today (which is just segmentation).
4
The How
One way to bring MyTwin to life, building on my previous essay on Large Customer Models, is through the creation of a Single Customer Model (SCM). This comprehensive model integrates multiple layers of intelligence:
- Enterprise-wide customer data aggregation and analysis
- Dynamic segmentation patterns and insights
- Complete product and service catalogues
- External adtech signals (search and social behaviours)
- Individual customer profiles (demographics, events, transactions)
- Zero-party data voluntarily shared by customers
Traditionally, this data remains fragmented across various platforms and databases. While CDPs (customer data platforms) or CDWs (customer data warehouses) attempt to unify this information, they often fall short of creating truly actionable intelligence.
The real breakthrough in making SCM a reality comes from the convergence of three generations of AI:
- Predictive AI for using pattern recognition and behavioural analysis to anticipate needs
- Generative AI enabling natural language interactions
- Agentic AI powering autonomous decision-making through tools like the Co-Marketer
When combined with innovations like NeoMails and AI Twins, these AI capabilities transform the SCM into a living, breathing entity. MyTwin becomes the multi-faceted manifestation of this intelligence – a conversational layer sitting atop the SCM, much like ChatGPT’s relationship to foundational LLMs. MyTwin can revolutionise every touchpoint in the customer-brand relationship:
- Predictive next-best-actions throughout the customer journey
- Dynamic recommendations across websites and apps
- Hyper-personalised communications across all channels
- Tailored microcontent in daily NeoMails
- Contextually relevant questions in SmartBlocks for zero-party data collection
- Precision advertising that feels like valuable content
- Curated cross-brand recommendations
The fundamental shift: moving from segments, campaigns, and generic journeys to dynamic, personalised experiences with generative journeys which adapt in real-time – more like Google Maps continuously recalculating routes than following directions on a paper map.
This represents uncharted territory for marketers. By harnessing next-generation AI, we can finally realise the vision articulated 30 years ago in ” in The One to One Future: Building Relationships One Customer at a Time” by Don Peppers and Martha Rogers. When we look back, the 2000-2025 era of marketing will be seen as an aberration – a period marked by massive AdWaste in the relentless pursuit of growth.
MyTwin enables the transformation from the AAA era (acquisition, acquisition, acquisition) to the OOO era (Only Once/Ones). In doing so, it helps marketers achieve the trifecta of success:
- Maximising customer lifetime value through deeper engagement
- Minimising acquisition costs by preventing customer dormancy
- Monetising attention through precision PII-based advertising
This marks the dawn of the NeoMartech/NeoAdtech era, where brands, not Big Tech, control the relationship with their customers. It promises an age of abundant profits as hundreds of billions in AdWaste are redirected to brand P&Ls, fundamentally reshaping the economics of digital marketing.
5
Example
I asked Claude to create a conversation between Ria and her eCommerce brand twin (MyTwin).
Ria: Hi MyTwin, I need some help with picking outfits for my upcoming business trip to Singapore next month.
Ria’s MyTwin: I know from your past purchases you prefer business casual that works in tropical weather. You usually choose navy and beige as base colors. How long is the trip?
Ria: It’s a 5-day trip, with mostly client meetings. But I also want to do some sightseeing over the weekend.
MyTwin: Given your schedule, I’d suggest 3 business outfits that can be mixed and matched, plus 2 casual ones. I notice the linen blazer you bought last summer worked well for you in similar weather. Would you like to build around that?
Ria: Yes, but I think I need new tops to go with it. The white one I usually pair it with is looking worn out.
MyTwin: I can help with that. Based on your style preferences and past purchases, here are a few suggestions:
- A sleeveless silk blouse in ivory (it’s more durable than the cotton one you had)
- A light blue wrap top that can work for both meetings and dinner
- A moisture-wicking fitted tee in navy for casual days
I’ve also noticed you often look for pieces that can transition from work to evening events.
Ria: That’s right. And I’m trying to build a more sustainable wardrobe.
MyTwin: I’ll adjust the recommendations to focus on quality pieces that last longer. Also, I know you usually wear size M, but for this brand’s wrap tops, several customers with your build found the S more flattering. Would you like to see both options?
Ria: Yes, please. And what about bottoms? I’ll need something comfortable for the weekend sightseeing.
MyTwin: Given Singapore’s climate and your preference for comfort, I’d recommend the breathable stretch trousers that just launched – similar to the black ones you loved last year but in a lighter fabric. They’ve received great reviews from other business travelers. For weekends, remember those linen culottes you saved to your wishlist? They’re now on sale and would pair perfectly with both the navy tee and wrap top.
Ria: Perfect! Can you also suggest some accessories that won’t take up too much luggage space?
MyTwin: Based on your minimalist accessory style, I’d recommend:
- The convertible crossbody bag you were eyeing last month (it’s back in stock in the tan color you prefer)
- A lightweight printed scarf that can dress up the basic tops
- Your existing pearl studs work perfectly for meetings
By the way, I notice you usually buy new walking shoes before big trips. Your last pair was purchased 8 months ago – would you like to see some options that are both business-appropriate and comfortable for sightseeing?
I also asked Claude to showcase how Ria’s shopping experience typically works now.
- Receives multiple promotional emails daily from the brand, most featuring items irrelevant to her needs
- Browses website, having to:
- Navigate through numerous category pages
- Filter by size, color, occasion repeatedly
- Read generic product descriptions
- Scroll through pages of irrelevant items
- Adds items to cart, unsure about:
- How pieces will work together
- Whether sizes will be consistent
- If items are appropriate for Singapore’s climate
- Quality and durability
- Gets retargeted with ads showing exact items she already purchased or decided against
- Receives generic post-purchase emails suggesting random products rather than complementary pieces
- Has to start the entire process again for different categories (shoes, accessories, etc.)
**
The contrast highlights how MyTwin transforms the experience from transactional and fragmented to conversational and cohesive, while saving time and reducing purchase anxiety. Instead of Ria adapting to the brand’s structure and processes, MyTwin adapts to Ria’s needs and preferences.
