My Take (focused on Marketing) – 1
Here is what I had written last year (Part 10):
2025 will mark the convergence of three transformative AI technologies that fundamentally reshape how brands connect with customers. The combination of Agentic AI (through AI Co-Marketers), AI Twins, and true N=1 personalisation will finally solve marketing’s persistent “Not for Me” problem – where generic messaging fails to resonate with individual customers.
AI Twins will evolve beyond basic segmentation to create personal digital companions for each customer. Starting with Adtech Twins built from public data, progressing to Madtech Twins incorporating marketing insights, and culminating in individual Singular Twins (MyTwins), these AI replicas will enable unprecedented understanding of customer needs and preferences.
The AI Co-Marketer will serve as the orchestration layer, using Agentic AI to coordinate across these Twins and create what I call “Generative Journeys” – dynamic customer paths that adapt in real-time like Google Maps recalculating routes. This combination will enable true N=1 personalisation at scale, where every interaction feels personally crafted for each customer.
Most importantly, this trinity of AI innovations will transform marketing from mass communication to individual conversation. Rather than bombarding customers with generic messages, brands will engage in meaningful dialogue through AI-powered personal companions. The result? Higher engagement, better retention, and dramatically lower customer acquisition costs as brands shift from endless acquisition to building lasting, profitable relationships.
This AI-powered transformation of customer engagement lies at the heart of what I call “NeoMarketing” – a revolutionary paradigm that moves beyond mass messaging and repeated acquisition to create genuine one-to-one relationships for lifelong retention at scale.
Last year, I wrote about the convergence of Agentic AI, AI Twins, and N=1 personalisation solving marketing’s persistent “Not for Me” problem. The direction was right; the timing, as always, was optimistic.
Marketing is indeed becoming agentic—but the transformation is unfolding across two parallel tracks: operational automation (Marketing Agents with a Co-Marketer running campaigns) and customer-facing intelligence (AI understanding each individual at unprecedented depth). The broad direction remains: “Department of One for Segment of One.”
This year, I want to go deeper on three forward-looking shifts I’m actively working on—each of which I believe will reshape marketing economics, customer understanding, and channel strategy over the 2026–2028 arc.
- Alpha: Value-Based Pricing Becomes the Default Expectation
In 2026, the martech procurement conversation starts to change. For the first time, a meaningful subset of CMOs and CFOs will ask a new question:
“Why are we paying for messages and modules when what we want is outcomes?”
The driver is simple: AI collapses the cost of execution. Copy, creatives, segmentation, variants, basic automation—all approach zero marginal cost. When everybody can generate 50 campaign variants in seconds, the differentiator is no longer production volume. The value migrates from “doing” to “delivering.”
That’s the opening for Alpha pricing—borrowed from finance, where Alpha means returns above the benchmark. Instead of paying for the capacity to send emails, brands pay for the incremental revenue those emails generate. Instead of licensing a CDP based on records stored, pricing reflects the lift in customer lifetime value.
The transition unfolds in three phases:
- Phase 1 (2026): Humans-as-a-Service. Growth engineers sit between the brand and the platform, committing to measurable lifts—reactivation rate, repeat purchase, contribution margin, retention. They use AI tools but take accountability for results.
- Phase 2 (2027): Agent-Assisted Delivery. AI agents take over operational execution—testing, orchestration, anomaly detection, next-best-actions—reducing cost-to-serve and raising confidence in performance commitments.
- Phase 3 (2028): Agent-to-Agent Performance Contracts. The vendor’s agent commits to outcomes, reports transparently, and continuously rebalances tactics to keep performance on track. Fully autonomous, fully accountable.
The prediction: The pricing unit shifts from messages sent and features licensed to retention lift and profit per customer. Vendors who cannot align economics will increasingly be treated as interchangeable utilities.
What to watch in 2026: More contracts with variable components, guarantees, “shared upside” structures, and CFO involvement in martech renewal decisions.
Why this matters for the Great Rebalancing: Alpha pricing aligns incentives between brands and vendors for the first time. When martech is priced on outcomes, the ROI of reducing churn becomes as visible as the ROI of acquiring new customers. The rebalancing from acquisition to retention begins when CFOs can see Alpha in their P&L.
- Artificial People: Consumer World Models Make Hyper-Personalisation Real
AI has made “personalisation” cheap. It has not yet made it true.
Most personalisation today is still an elaborate form of rules and propensity scores—constrained by sparse signals, weak understanding, and the fundamental limitation of treating customers as rows in a database rather than as living, deciding, evolving humans.
The next leap comes when brands stop modelling “customers” and start modelling customer behaviour. That’s where Artificial People—what I call consumer world models—becomes the foundational primitive.
A World Model is a physics engine for human behaviour. Just as a self-driving car’s world model simulates road conditions, obstacles, and other drivers, a consumer world model simulates:
- Decision dynamics: How this person weighs price vs. convenience vs. brand affinity
- Temporal patterns: When they’re receptive vs. resistant to outreach
- Context sensitivity: How behaviour shifts across channels, devices, moods, life stages
- Influence networks: Who and what shapes their preferences
This enables the industry to move from:
Data → Segments → Campaigns
to:
Models → Predictions → Conversations
Artificial People creates conversable, evolving archetypes that map to individual customers and update continuously. Three things become possible that weren’t before:
- Stable meaning from messy signals. Archetypes give coherence to sparse and noisy data—you don’t need perfect information to understand someone.
- Real-time intent and timing. The model predicts not just what a customer might want, but when they’re most receptive and why.
- A single “customer brain” across channels. Email, onsite, app, WhatsApp, support, and commerce all draw from one evolving representation—no more channel silos with conflicting views of the same person.
The prediction: The martech stack shifts from “event-stream + rules” to “world-model + agents.” CDPs don’t disappear, but they become plumbing. The customer model layer becomes the strategic asset—and the new moat.
What to watch in 2026: Vendors talking less about segmentation and more about “customer model layers,” memory, embeddings, and continuously updating representations.
The collision with agentic commerce: Here’s where it gets interesting. In a world where AI agents shop on behalf of consumers, brands aren’t just marketing to humans—they’re marketing to the models that represent humans. The consumer’s AI agent has its own understanding of their preferences. The brand’s Artificial Person has its own simulation. Marketing becomes a negotiation between world models. The brands that win will be those whose models most accurately simulate customer behaviour—and can “speak” to the customer’s own AI agent in terms it understands.
- NeoMails: The Inbox Becomes an Attention and Monetisation Surface
Email has been treated as a cost centre—a channel brands pay to access. Open rates reflect fatigue, not engagement. The inbox is polluted by volume, and every marketer’s response to declining performance is to send more.
In 2026, two forces collide to change this:
- The Attention Crisis: As AI agents and AI-generated content explode, human attention becomes scarcer and more expensive. The moments when a human actually reads something become precious.
- The Owned Media Realisation: Brands will try harder to reclaim engagement from rented platforms (Meta, Google) by creating repeatable rituals on owned channels. Email is the last direct line to consumers who have explicitly opted in.
NeoMails reimagines email through a simple inversion: instead of brands paying to send emails that customers ignore, brands create emails so valuable that advertisers pay to be included—and customers want to open.
The model:
- ZeroCPM Marketing Emails: The brand’s email to its customer costs nothing to send because it’s subsidised by relevant, non-competitive advertising embedded within.
- Attention as Currency: The customer’s demonstrated attention—opens, reads, clicks, time spent—becomes a monetisable asset. Brands share that value with customers (through rewards, exclusive content, better offers) or reinvest it in better experiences.
- The Inbox as Daily Utility: The strongest email programs will look less like campaigns and more like daily utilities—content, games, recommendations, micro-experiences, interactive moments. They become habit-forming, not interruptive.
The prediction: The winning owned-channel programs will look less like promotional calendars and more like media products with recurring formats and rituals. Email gets reborn as a product, not a channel. And new monetisation structures—not increasing ESP bills—will fund attention creation.
What to watch in 2026: More interactive email, more embedded actions, more inbox-native experiences, more experimentation with ad-supported owned media.
Why this works in an agentic world: As AI agents filter and triage communications on behalf of consumers, only high-value, high-relevance messages will get through. NeoMails are designed to pass that filter—they deliver genuine value, not noise. The brands that master this will be the ones whose emails AI agents allow into the consumer’s attention stream.
