Imagining Meridian: A Proprietary Model for Guaranteed Outcomes

Published February 11-22, 2026

The Tool Paradox

Marketing has never had more capability than it has today.

A modern customer engagement stack can send billions of messages, personalise at the level of the individual, automate journeys across dozens of channels, and optimise product discovery in real time. AI has made what used to be expensive — copy variations, audience creation, creative iteration, experimentation — almost free. The pipes are everywhere. The intelligence is abundant. The dashboards glitter with real-time analytics that would have seemed magical a decade ago.

And yet the defining outcome that matters most to a business has barely moved.

Customers still drift away at scale. Brands still lose the quiet majority without noticing until the numbers surface in the next quarter’s report. Eighty percent of engaged customers vanish within ninety days — not through dramatic unsubscribes or angry complaints, but through the slow fade of ignored messages, unopened emails, and sessions that never happen. They don’t leave. They simply stop showing up.

And then brands do the most irrational thing in business: they pay again to buy back the same customers they already had.

This is the core of the $500 billion AdWaste crisis. It is not an inefficiency to be optimised. It is a structural failure to be replaced. A global, compounding “reacquisition tax” paid to platforms because brands cannot reliably retain and reactivate the customers already sitting in their databases. If this were merely inefficiency, the market would have corrected it by now. Two decades of martech innovation have not corrected it. Which means the problem is not a lack of tools.

The problem is a lack of accountability.

For twenty years, marketing technology has been sold as capability — more features, more channels, more automation, more AI. The buyer has almost always been the marketing function. The pricing model has almost always been based on inputs: messages sent, seats licensed, events stored, journeys created, records managed. The vendor gets paid for access. The brand hopes for impact.

When martech works, the brand wins. When martech fails, the brand loses. Either way, the vendor invoices on the first of the month.

This misalignment is not a bug in the system. It is the system. And until the system changes, the outcomes will not change either. You cannot expect accountability from a business model that does not require it. You cannot expect retention outcomes from vendors whose revenue is divorced from whether customers stay or leave.

The capability explosion has not produced an outcome revolution because capability and accountability are different things. One is about what you can do. The other is about what you must deliver. Marketing technology has been built almost entirely around the first. The second has been left to hope, effort, and quarterly post-mortems that explain why this time was different.

Consider the asymmetry. When a brand’s customers churn, the brand loses their lifetime value, pays reacquisition costs to win them back, and absorbs the margin compression of competing for attention they once owned for free. When a brand’s customers churn, the martech vendor loses nothing. The contract continues. The platform fee arrives. The customer success manager schedules a call to discuss “optimisation opportunities.”

This is not a criticism of any particular vendor. It is a description of how the category was built. Enterprise software pricing — seats, usage, modules — was inherited from an era when software was a tool and outcomes were the buyer’s responsibility. That model made sense for productivity applications. It makes less sense for systems whose entire purpose is to produce measurable business results.

The strange truth is that marketing technology is sold as a strategic investment but priced as a utility. Electricity companies do not promise business outcomes. They promise kilowatt-hours. Martech vendors promise transformation but invoice for throughput. The mismatch is so normalised that questioning it feels naive.

But what if the question were asked differently?

What if marketing technology had been built like a performance category from the beginning? What if it had been priced like underwriting, not utilities? What if vendors had been accountable like portfolio managers, not software suppliers? What if “Never Lose Customers” were a contractual commitment rather than a conference keynote?

That thought leads to a different model entirely. Not better tools. Not smarter dashboards. Not more sophisticated segmentation. A different accountability structure that aligns vendor economics with brand outcomes — so that the people building the intelligence have the same incentives as the people deploying it.

This essay imagines that model. A proprietary marketing AI built to deliver measurable retention and profit uplift, priced on outcomes, and sold not as a toolbox to a department but as a performance engine to executives. A model called Meridian: a reference line that enables precise positioning. Meridian’s promise is not better tools. It is Alpha — verifiable uplift in customer profit metrics — and an accountability contract that makes “Never Lose Customers” more than a slogan.

PS: Here is a longer explanation on how martech lost the game it was meant to win.

Martech wasn’t beaten by adtech. It was abandoned by design:

Abandoned by pricing that rewarded activity over outcomes
Abandoned by complexity that exceeded human capacity to operate
Abandoned by personalisation promises that collapsed into stereotypes
Abandoned by measurement frameworks that couldn’t see attention decay
Abandoned by channel abuse that destroyed owned attention
Abandoned by strategic blindness that ignored 80% of customers
Abandoned by ease of alternatives that required no expertise
Abandoned by attribution that couldn’t distinguish acquisition from reacquisition

Why DIY Hits a Ceiling

The industry’s default answer to retention failure has been familiar for two decades: buy better tools and hire smarter people. That answer is not wrong. It is incomplete. It helps, up to a point. Then it hits a ceiling.

This is not a criticism of marketers. It is a recognition of structural constraints that even the best marketing teams cannot overcome. The constraints are not about talent or effort. They are about the operating model itself — the way marketing organisations are structured, measured, funded, and held accountable. Understanding these constraints is essential before proposing an alternative.

The first constraint is feature underutilisation. Studies consistently show that 60-65% of martech platform capabilities go unused. This is not because marketing teams are incompetent. It is because modern platforms are complex systems whose power lies in configuration, data architecture, experimentation discipline, and constant iteration. The work is never finished. The learning curve is perpetual. Staff turnover means the organisation repeatedly climbs the same curve with different people.

Buying a sophisticated platform and using it like a bulk messaging system is like buying a Bloomberg terminal and checking only stock prices. Or a Ferrari without a driver. The capability exists. The expertise to exploit it does not persist. You own the tool. You do not own the outcome.

The second constraint is the human bottleneck. Even with AI assistance, the operational surface area of modern marketing is enormous: audiences and segments, journeys and triggers, creative variants and messaging frameworks, analytics and measurement, experimentation and learning transfer, cross-channel coordination, compliance and governance and approvals. Each of these domains requires attention, judgement, and iteration.

Humans remain the rate limiter. People leave. Teams restructure. Institutional knowledge disappears. The reasoning behind previous decisions evaporates when the person who made them moves on. Organisations find themselves repeating the same learning cycles every eighteen months with different faces, rediscovering insights that were documented in a deck no one can find.

Even brilliant marketers have bandwidth constraints. A team that could execute flawlessly on one channel struggles when asked to coordinate five. A strategist who could design a perfect journey for one segment cannot design perfect journeys for fifty. The gap between what is possible and what is practical widens as complexity increases. At scale, the system becomes fragile — dependent on individuals who cannot be replicated and processes that cannot be sustained.

The third constraint is campaign-mode thinking. Most marketing organisations operate in bursts: launch, measure, report, pause, repeat. Planning cycles are quarterly. Optimisation happens in retrospectives. Learning is captured in presentations that are reviewed once and never referenced again. Improvements come as step changes — a new platform, a new agency, a new strategy — rather than continuous compounding.

But customer drift is continuous. Attention decay is continuous. Competitive pressure is continuous. The operating model is mismatched to reality. Customers do not disengage in quarterly cycles. They disengage in daily moments of irrelevance, accumulating into weekly habits of ignoring, crystallising into monthly patterns of absence. By the time the quarterly review surfaces the problem, the customers are gone.

The fourth constraint is cost-centre economics. Marketing is almost always run as a budget function, not a profit-and-loss centre. This changes behaviour in profound ways. Budget functions optimise for spend justification. They measure activity because activity is easy to report. They celebrate outputs — campaigns launched, emails sent, impressions delivered — because outputs are controllable. P&L functions optimise for outcomes because outcomes are what matter. They ask different questions: not “what did we do?” but “what did we earn?”

When martech spend is not directly accountable to incremental customer profit, the system naturally drifts toward vanity metrics and comfort metrics. Open rates improve while retention rates stagnate. Click-through rates edge upward while customer lifetime value stays flat. The dashboard looks healthy. The business does not.

The fifth constraint is the adoption paradox. The future concepts that could transform retention — Context Graphs, BrandTwins, N=1 decisioning, agentic orchestration — are not features you turn on. They are new mental models. They require different ways of thinking about customers, data, decisions, and measurement. And organisations struggle to implement what they do not fully grasp.

By the time most companies build internal capability to deploy these concepts, the competitive window has closed and the market has moved again. The early adopters compound their advantage. The laggards fall further behind. The gap widens not because of technology access — the tools are available to everyone — but because of execution capability, which is not.

The aggregate effect of these five constraints is stark: even the best tools require perfect execution to deliver consistent outcomes. Perfect execution at scale is structurally impossible for human teams operating within quarterly planning cycles, annual budget allocations, eighteen-month staff turnover, and cost-centre accountability.

This is not defeatism. It is realism. The DIY model can improve retention. The DIY model cannot reliably guarantee retention. The gap between “improved” and “guaranteed” is precisely the space where a different model becomes necessary.

The question is not whether to abandon DIY. Many organisations want control. They want internal capability. They want ownership of their marketing operations. For them, the platform-plus-agents model is the right starting point.

The question is what to offer organisations that want outcomes instead of tools — executives who do not want to build internal capability but want to buy external accountability. For them, a second path is needed. Not “do it better” but “do it differently.”

Two Paths to Never Lose Customers

If the mission is “Never Lose Customers,” there are two ways to pursue it. Both are legitimate. Both serve different organisational needs. Both can coexist within the same company. But they represent fundamentally different relationships between the buyer and the provider.

The first path is the traditional model, upgraded with AI. Call it the Do-It-Yourself path — or more precisely, Agentic Marketing.

This means a customer engagement platform to orchestrate channels and journeys: email, push, in-app, SMS, WhatsApp, RCS, web personalisation. It means a discovery layer to optimise search, browse, and recommendations — turning the on-site experience into a responsive surface rather than a static catalogue. And it means a suite of marketing agents (M-Agents) to amplify the team’s capabilities.

These agents are not chatbots. They are functional specialists embedded in the workflow:

An Insights Agent that surfaces patterns humans miss — anomalies in engagement, early signals of churn, unexpected correlations between behaviour and outcome. A Segmentation Agent that creates dynamic cohorts based on real-time behaviour rather than static attributes, continuously refining audiences as customers act. A Content Agent that generates and tests variations at scale, producing personalised messaging that would take human teams weeks to create manually. A Shopping Agent that optimises purchase paths, reducing friction and increasing conversion through intelligent nudges. A Merchandising Agent that balances inventory levels, margin targets, return rates, and availability constraints to surface the right products at the right time. And an Orchestrator — a Co-Marketer — that coordinates across all these specialists, ensuring actions are coherent rather than contradictory.

This is powerful. This is what gets sold to CMOs and marketing operations leaders. This is priced as SaaS — monthly or annual fees for access to capabilities. The promise is straightforward: “We give your team better tools. You execute. You own the outcome.”

For many organisations, this is the right path. They want control over their marketing operations. They want to build internal capability that compounds over time. They want ownership of the strategy, the execution, and the learning. The platform-plus-agents model gives them leverage without giving up autonomy.

But the Agentic Marketing path leaves the biggest question untouched: who is accountable when customers leave?

The tools are accountable for working as specified. The vendor is accountable for uptime, deliverability, feature releases. But the outcome — whether customers stay, whether lifetime value grows, whether AdWaste shrinks — remains the brand’s problem. If the tools are used well, the brand benefits. If the tools are used poorly, or used well but against structural constraints that doom the effort, the brand absorbs the loss.

This is where the second path emerges. Call it the Done-For-You path — or more precisely, NeoMarketing.

NeoMarketing uses agentic intelligence — but with a difference. The agents don’t just assist. They’re accountable.

The proposition is different in kind, not just degree: “We take accountability for retention and customer profit uplift. You measure the delta. You pay only when we deliver.”

Same underlying infrastructure. Same Context Graphs, same agent capabilities, same channel orchestration. But a different accountability model. A different buyer. A different pricing structure. And therefore, different results.

NeoMarketing is built on two components, matched to customer value:

Meridian operates on Best customers — the top-value segment where deep intelligence is economically justified. These are customers whose lifetime value is high enough that sophisticated N=1 treatment generates meaningful return on the compute and attention invested. For Best customers, every interaction matters. Getting it wrong is expensive. Getting it right compounds. Meridian helps “MAX the LTV.”
NEO operates on Rest and Test customers — the larger pool where attention recovery must happen at scale. Rest customers are the quietly disengaging middle — still technically active but fading. Test customers are already lapsed — gone but potentially recoverable. For these segments, the goal is not deep intelligence but systematic recovery: rebuild the habit of engagement on owned channels like email before absence becomes permanent. NEO helps “ZERO the CAC.”

The critical distinction is not just segmentation. It is the buyer.

Agentic Marketing is sold to CMOs and marketing operations leaders. They evaluate features, integrations, ease of use, and support. They manage implementation. They own the execution. Their success is measured by how well they use the tools.

NeoMarketing is sold to CEOs and CFOs. They evaluate outcomes: retention improvement, profit uplift, reduction in reacquisition costs. They do not want to manage implementation. They want to measure results. Their success is measured by what appears in the P&L, not by how sophisticated the marketing operation has become.

This difference in buyer changes everything.

CMOs buy capability because their job is to build and execute marketing programmes. They need tools that make their teams more effective. They are evaluated on marketing metrics.

CEOs and CFOs buy outcomes because their job is to grow the business profitably. They do not care whether the outcome is achieved through sophisticated AI or through a thousand monkeys with typewriters. They care whether customer retention improves, whether AdWaste decreases, and whether the investment generates measurable return.

The Agentic Marketing path answers the CMO’s question: “How do I execute better?”

NeoMarketing answers the CEO’s question: “Why do I keep paying to reacquire customers I already had?”

Both questions are valid. Both deserve answers. But they are different questions, and they require different commercial structures.

Agentic Marketing is priced on inputs — what you can access. NeoMarketing is priced on outputs — what gets delivered. Agentic Marketing transfers risk to the brand — if execution fails, the brand loses. NeoMarketing transfers risk to the provider — if outcomes fail, the provider does not get paid.

This is not a small shift. It is a category redefinition.

M-Agents help marketers execute retention. NeoMarketing guarantees retention outcomes. The mission is the same — Never Lose Customers. The accountability is opposite.

To understand how that guarantee becomes possible — how Meridian MAXes LTV and NEO ZEROs CAC — we need to look inside Meridian first.

The Proprietary Intelligence – 1

Before defining what Meridian is, it is worth being precise about what it is not. The market is saturated with “AI-powered marketing” claims. Most of them describe incremental improvements to existing approaches. Meridian describes something structurally different.

Meridian is not a generic large language model wrapped in a marketing interface. Generic LLMs are trained on public internet text. They can write. They can summarise. They can generate variations. What they cannot do is reason about your specific customers, your specific products, your specific margin structures, your specific competitive dynamics, your specific history of what worked and what failed. Wrapping a general-purpose AI in a marketing user interface produces general-purpose results — which is to say, mediocre results that any competitor can replicate with the same wrapper.

Meridian is not journey automation with better copy. Most “AI-powered” marketing tools take the existing campaign-mode paradigm and accelerate it: more segments, more variations, more journeys, more tests. They optimise within the existing model without questioning whether the model itself is the problem. Faster mediocrity is still mediocrity. More sophisticated segmentation is still segmentation. If the underlying approach loses eighty percent of engaged customers per quarter, making that approach faster does not solve the problem.

Meridian is not a dashboard or insights layer. The industry has no shortage of analytics that tell marketers what happened. Dashboards are excellent at describing the past. They are useless at changing the future. Insights without action create sophisticated paralysis — teams that know exactly how they are failing but lack the operational capacity to do anything about it. Meridian does not primarily report. Meridian acts.

So what is Meridian?

Meridian is a proprietary decisioning model built on Context Graphs that creates a BrandTwin for each valuable customer, orchestrated by agents that act continuously and autonomously, and priced on the outcomes it delivers.

That definition has four layers. Each layer matters. Each layer builds on the one before.

Layer 1: Context Graphs as Substrate

[See Context Graphs: The Missing Layer Between Data and Action
2026 AI and Marketing and NEO’s Context Graph Advantage: Turning Customer Drift into Compounding Attention]

Most customer systems store data. Meridian requires something different: structured intelligence optimised for reasoning and action.

The distinction matters. A traditional customer data platform stores facts: this customer bought this product on this date, opened this email at this time, visited this page in this session. Facts are necessary but insufficient. Facts tell you what happened. They do not tell you what it means or what to do about it.

A Context Graph is not a dashboard. It is not a data warehouse. It is not merely a knowledge graph with marketing data. A knowledge graph answers the question: what do we know? A Context Graph answers a different question: what matters now, and why?

The architecture has three interlocking components.

The Customer Context Graph captures state and trajectory. It knows not just where a customer is but where they are heading. This includes identity and relationship history across channels, preferences both explicit and inferred, engagement rhythms and channel behaviours, lifecycle stage signals, value trajectory including lifetime value and contribution margin and churn risk, interaction history across messages and sessions and transactions, and crucially, confidence scores that quantify how certain the system is about each inference. Not all knowledge is equal. The system must know what it knows and what it merely guesses.

The Product Context Graph captures relationships and economics. It understands catalogue structure — which products are substitutes, which are complements, which are accessories, which are gateways to larger purchases. It tracks inventory and availability in real time. It knows margin and unit economics at the SKU level. It monitors return rates and dissatisfaction signals. It understands seasonality and demand patterns. This layer ensures that recommendations are not just relevant to the customer but profitable to the business and actually available to fulfil.

The Decision Trace Graph captures what most systems entirely ignore: why decisions were made. Not just that a discount was offered, but what signals triggered the intervention, what alternatives were considered, what hypothesis was being tested, what the outcome was, and what precedent was established for future similar situations.

This is the critical differentiator. Most enterprise software manages the “state clock” well — what is true right now. Almost nothing manages the “event clock” — what happened, in what order, with what reasoning, producing what result. When humans were the decision layer, the “why” lived in people’s heads. Institutional knowledge was reconstructed through conversation and experience. But AI agents do not have heads. They cannot remember why a decision was made unless that reasoning is explicitly stored. Without decision traces, every choice starts from scratch. The system cannot learn from its own experience. It cannot transfer insights from one customer to another. It cannot compound intelligence over time.

Meridian’s Context Graph is built to hold both state and reason. The state tells you where you are. The reason tells you how you got there and what to do next.

The Proprietary Intelligence – 2

Layer 2: BrandTwins as Representation

A BrandTwin is an N=1 counterpart for each Best customer. Not a profile. Not a segment label. Not a collection of attributes in a database row. A BrandTwin is a continuously updated model of a single customer’s relationship with the brand.

The BrandTwin knows what the customer values — not demographically but behaviourally, based on what they choose, what they ignore, what they return, what they recommend. It knows what the customer is drifting away from — the early signals of disengagement that precede the visible symptoms. It knows what pulls the customer back — which messages resonate, which offers convert, which experiences rebuild the habit of engagement. It knows what the brand has promised — explicitly through communications and implicitly through prior treatment — and what the customer therefore expects next.

Crucially, each BrandTwin carries a micro P&L. It tracks revenue generated by this specific customer. It tracks cost invested — in incentives, in service, in interventions, in attention. It calculates contribution margin impact. It estimates retention probability. It identifies the next-best investment opportunity — the action most likely to generate incremental value given the current state and trajectory.

This is where Meridian begins to feel less like marketing and more like capital allocation. The BrandTwin’s job is not to maximise clicks or opens or even conversions in isolation. Its job is to maximise relationship value over time, within constraints. It must balance short-term conversion pressure against long-term trust. It must recognise when an aggressive offer would generate a transaction but erode the relationship. It must know when not to act — when silence is more valuable than another message.

The BrandTwin becomes, in effect, the customer’s advocate inside the system. It represents the customer’s interests because representing those interests is how value is maximised. A customer who feels understood stays longer, buys more, forgives occasional failures, and recommends to others. A customer who feels exploited leaves at the first opportunity.

This is not altruism. It is aligned economics. The BrandTwin protects the customer because protecting the customer protects the revenue stream.

Layer 3: Agents as Execution

Meridian’s intelligence is meaningless without execution. Context Graphs can hold perfect understanding. BrandTwins can model customers with exquisite precision. None of it matters if actions do not follow.

This is where agents operate — not as chatbots or assistants but as autonomous operators running continuous micro-experiments across the customer base:

Adjusting content, timing, and cadence for each customer based on predicted responsiveness.
Choosing channels based on where this specific customer is most likely to engage, not where the average customer engages.
Selecting offers based on margin and elasticity — ensuring that discounts go to customers who need them to convert, not to customers who would have converted anyway.
Triggering service interventions when trust risk rises — a proactive call, a surprise upgrade, a personal note — before the customer reaches the threshold of departure.
Coordinating on-site discovery experiences with off-site messaging — so that the email a customer receives connects to the products they see when they visit.
Detecting drift earlier than human reporting cycles can surface it — recognising the pattern of disengagement while intervention is still possible.

Instead of campaign bursts, Meridian runs a living system. Hundreds of decisions per day, per customer base, guided by a consistent model of what matters. This is not “set and forget” automation. It is genuine agency — the capacity to perceive, decide, and act based on goals rather than rules.

The difference matters. Rule-based automation executes the same action whenever a trigger fires, regardless of context. Agent-based intelligence evaluates context, considers alternatives, estimates outcomes, and selects the action most likely to achieve the goal. The first is mechanical. The second is adaptive.

Layer 4: TwinFactory as Scale

A single BrandTwin is interesting. A million BrandTwins are infrastructure.

TwinFactory is the machinery that creates, maintains, and updates BrandTwins at scale. It ingests signals continuously from every touchpoint. It updates Context Graphs and confidence scores as new information arrives. It replays decision traces to transfer learning from one customer’s experience to similar customers’ predictions. It keeps the system consistent as products change, inventory shifts, pricing evolves, and customer behaviour adapts.

This is what transforms an AI concept into an operational system. The challenge of N=1 personalisation has always been scale. How do you deliver individual intelligence to millions of customers without employing millions of marketers? TwinFactory is the answer: industrialised twin creation that is systematic, repeatable, and autonomous.

The navigational metaphor

A meridian is a reference line used for precise positioning. Sailors used meridians to know exactly where they were on the globe — not approximately, not “somewhere in this region,” but precisely. The meridian did not move. It provided a fixed reference against which position could be calculated.

Meridian positions each customer precisely in their journey. It knows not just that a customer is drifting but why they are drifting, where they are drifting toward, and what will arrest the drift. It uses signals others cannot see — the subtle patterns in behaviour that precede visible symptoms. It makes decisions others cannot justify — interventions that seem counterintuitive until you understand the underlying model. It delivers outcomes others cannot guarantee — because it is built for guarantee, not hope.

And like any powerful proprietary model, Meridian’s promise is not that you will understand it. Meridian’s promise is that you will measure what it delivers.

NEO — Attention Recovery at Scale – 1

Meridian’s depth is economically justified only for Best customers — those whose lifetime value warrants the investment in N=1 intelligence. But Best customers are typically only twenty percent of the base. What about the other eighty percent?

This is where NEO operates: the attention recovery system for Rest and Test customers.

The segmentation matters. Rest customers are the quietly disengaging middle. They have not unsubscribed. They have not complained. They have not dramatically departed. They have simply faded. They open fewer emails. They visit less frequently. They buy less often. They are technically “active” by most measurement systems but behaviourally absent. They are drifting toward Test status without anyone noticing because the drift is gradual and the dashboards are not built to detect gradual.

Test customers have already lapsed. They have not engaged in months. They may still be in the database, but they have mentally moved on. They are the ones brands pay Google and Meta to “reach” through programmatic advertising — paying acquisition costs for customers they already own but can no longer access through owned channels.

Together, Rest and Test represent the bulk of the AdWaste crisis. Rest customers are the leak. Test customers are the flood. Fixing the leak prevents the flood. But both require intervention.

The challenge is that deep intelligence — BrandTwins, full Context Graphs, autonomous agents optimising individual journeys — is not economically viable for customers whose value is uncertain or low. The compute cost exceeds the expected return. The operational complexity exceeds the potential benefit. A different approach is needed: one that recovers attention at scale rather than optimising relationships individually.

This requires recognising a truth that most martech systems ignore: email and messaging channels are not broken. They are magnetless.

The pipes work. Deliverability is solvable. Technical capability is abundant. What is missing is a reason for customers to engage. Most brand emails are interruptions — requests for attention that provide no value in return. They ask customers to look at products, consider offers, read announcements. They take without giving. And customers, rationally, stop responding.

NEO creates magnets.

NeoMails for the Rest

NeoMails targets Rest customers — the quietly disengaging — with daily, habit-forming emails designed not as campaigns but as relationship rituals.

The model is simple: utility creates habit, habit creates relationship, relationship creates value.

A NeoMail arrives every day at a predictable time. It takes 60 seconds to consume. It provides genuine utility: weather, headlines, local information, puzzles, horoscopes, curated content — wrapped around brand elements that feel like service rather than selling. The customer opens not because they want to buy something but because they want the utility. The brand earns attention by providing value first.

This inverts the traditional email model. Traditional emails ask: “How do we get customers to look at what we want to show them?” NeoMails ask: “How do we give customers something valuable enough that they want to open every day?” The first approach treats attention as something to be captured. The second treats attention as something to be earned.

NeoMails introduce APUs — Attention Processing Units — as the infrastructure for transforming email from a broadcast channel into an attention economy. APUs have four components:

Magnets capture attention through interactive content — polls, games, utilities.
ActionAds monetise attention through relevant offers served while engagement is active.
Mu (Atomic Rewards) is the attention currency customers earn for opening, clicking, completing actions, and providing feedback.
Ledger tracks Mu accumulation and enables redemption — discounts, exclusive access, charitable donations, raffle entries for cash prizes.

The exchange is transparent. The customer understands that their attention has value and the brand is willing to compensate for it.

This is not gamification gimmickry. It is honest economics. Attention is scarce. Brands compete for it. Customers who grant attention deserve compensation. Mu makes the implicit exchange explicit — a visible currency that accumulates with every interaction, creating a micro-economy where both parties benefit from continued engagement.

NEO — Attention Recovery at Scale – 2

NeoBoost for Preventing Drift

NeoBoost is an attention layer that embeds into existing email programmes from any platform. It does not require migration. It does not require replacing the current ESP. It simply adds the attention mechanics of the APUs — the micro-rewards, the interactive elements, the habit-formation triggers — on top of whatever email infrastructure already exists.

This matters because platform switching is expensive and risky. Many organisations are locked into multi-year contracts, deeply integrated with existing systems, or simply unwilling to undertake the disruption of migration. NeoBoost offers the attention benefits without the switching costs.

The mechanism is straightforward: embed APU-earning interactions into existing email templates. Turn passive consumption into active engagement. Give customers a reason to interact rather than just read. The existing platform handles delivery and tracking. NeoBoost handles the attention economics.

NeoNet for the Already Lapsed

Test customers present a harder problem. They have stopped engaging with the brand’s owned channels. Emails go unopened. Push notifications are ignored. The brand has lost access.

The traditional solution is programmatic advertising: pay Google or Meta to reach these customers through ad inventory across the web. This works, but it is expensive. Programmatic CPMs are high. Targeting is probabilistic. The brand pays acquisition-level costs for customers it already “owns” in a database sense but cannot reach in a practical sense.

NeoNet offers an alternative: a cooperative identity network that allows brands to reach lapsed customers through other brands’ authenticated channels.

The logic is simple. The customer who stopped opening Brand A’s emails still opens Brand B’s. They are not disengaged from email. They are disengaged from Brand A. If Brand A could reach them through Brand B’s channel — with appropriate consent and compensation — the reacquisition problem becomes a coordination problem rather than an advertising problem.

NeoNet enables this coordination. Brands contribute authenticated reach to a shared network. They can access other brands’ engaged audiences for customers they have lost. The targeting is deterministic — based on actual identity, not probabilistic matching. The cost is a fraction of programmatic rates. The customer experience is better because the message arrives through a channel they already trust.

This is “Never Pay Twice” made operational. Why pay Meta fifteen dollars CPM to probabilistically reach someone in your database when you could reach them deterministically through NeoNet for a dollar fifty?

The shared substrate

Meridian and NEO run on the same Context Graph infrastructure. The difference is resolution.

Meridian uses the full individual trajectory — every signal, every decision trace, every nuance of a single customer’s evolving relationship. This depth is justified for Best customers whose value warrants the investment.

NEO uses cohort-level patterns — what works for customers who look like this, behave like this, and are at this stage of disengagement. This breadth is appropriate for Rest and Test customers whose individual value does not yet justify deep intelligence but whose aggregate value is enormous.

Same infrastructure. Different application. Matched to economic reality.

Together, Meridian and NEO form a complete retention architecture. Deep intelligence for those who have earned it through demonstrated value. Scalable recovery for those who have not — yet.

But even intelligence plus execution does not guarantee outcomes. Economics do.

The Alpha Contract

The real innovation in Meridian and NEO is not only the technology. It is the commercial structure.

The fundamental misalignment in marketing technology is that vendors get paid regardless of results. The contract specifies access to capabilities. The invoice arrives whether customers stay or leave. The brand bears the outcome risk. The vendor bears only the operational risk of keeping the platform running.

This structure made sense when software was genuinely a tool — a productivity enhancer whose value depended on how the buyer used it. Word processors do not guarantee good writing. Spreadsheets do not guarantee accurate analysis. The tool provides capability. The user provides skill. Outcomes depend on the combination.

But marketing technology is sold as something more than a tool. It is sold as a system for producing business results. Vendors promise transformation, competitive advantage, customer intimacy, personalisation at scale. The sales pitch is about outcomes. The contract is about inputs. The gap between promise and structure is the space where accountability disappears.

Alpha pricing closes that gap.

The model has three components:

Beta is the baseline — the revenues and retention the brand would have generated anyway through normal operations. Think of it like market returns in investing. This is the benchmark against which performance is measured.
Alpha is the incremental uplift above that baseline — the additional retention, revenue, and profit that NeoMarketing delivers beyond what would have happened without it. This is what gets measured. This is what gets paid.
Carry is the long-term component — participation in sustained improvement over time. Like private equity carry, it rewards durable value creation rather than short-term spikes.

The brand’s fixed cost is zero. All infrastructure, compute, and operational costs are absorbed by the NeoMarketing vendor. The brand pays only a share of the Alpha — the verified uplift above baseline. No uplift, no payment.

The philosophy is simple: we invest first, we deliver results, we get paid only when you earn more than you would have without us. We win only when you win. If there is no Alpha, there is no Carry. The vendor’s economics depend on delivering what was promised.

This structure transforms the conversation.

The traditional martech sales process involves feature comparisons, integration requirements, pricing negotiations, and implementation timelines. The buyer evaluates capability. The seller promises transformation. Everyone hopes for the best.

Alpha pricing changes the questions. The buyer asks: “What uplift are you committing to? How will it be measured? What happens if you miss?” The seller asks: “What does your current retention look like? What is the baseline we are improving against? What access do we need to deliver results?”

The conversation shifts from “what can this do?” to “what will this deliver?” Capability becomes a means rather than an end. Outcomes become the subject of negotiation rather than the object of hope.

The pitch to CEOs and CFOs becomes clean: “Give us ten percent of your customers for ninety days. We will run Meridian and NEO-driven interventions across your owned channels. You pay only on incremental profit uplift versus a matched control group. Stop anytime — but pay through the average purchase cycle so measurement captures delayed conversions.”

This structure does four things immediately:

It removes risk. The brand cannot lose money on the pilot. If there is no uplift, there is no payment. The worst case is that the brand learns something about its customer base and loses nothing but time.
It signals confidence. Offering Alpha pricing is a statement. It says: we believe in our model enough to bet our revenue on it. Vendors who refuse outcome-based structures are implicitly communicating something about their own confidence in their products.
It fits executive mental models. This is not marketing-speak about engagement scores and brand awareness indices. It is profit uplift — measurable, attributable, defensible in a board meeting, explicable to investors.
It creates natural expansion. Success with ten percent leads to twenty-five percent, leads to fifty percent, leads to full deployment. The pilot is the proof. The proof is the sale. No one needs to be convinced by presentations when results speak for themselves.

A useful parallel exists in finance. The best quantitative funds did not win by explaining their models. They won by compounding results. Renaissance Technologies delivered 39% annual returns (after fees) for decades without ever publishing their methodology. Investors did not need to understand the model. They needed to measure the returns.

Meridian brings similar discipline to marketing: proprietary intelligence, outcome accountability, measurement over explanation. The model is complex. The contract is simple. Pay for what you get.

Answering the Objections

Outcome-based marketing systems trigger predictable objections. They should. Healthy scepticism is part of sound executive decision-making. The objections deserve direct answers.

“Attribution will be disputed.”

This is the most common concern and the most easily resolved.

Meridian does not rely on last-touch attribution — the flawed model that credits the final click while ignoring everything that preceded it. Last-touch attribution creates endless disputes because it is fundamentally arbitrary. Why credit the last touch and not the first? Why credit any single touch when the customer journey involved dozens?

Meridian relies on uplift measurement. The methodology is straightforward: create two statistically identical populations. One receives Meridian-driven interventions. One receives business-as-usual treatment. Measure the difference in outcomes — retention, revenue, profit — between the two groups. That difference is the uplift. That uplift is what gets paid.

This is incrementality testing. It is the same methodology pharmaceutical companies use to measure drug efficacy. It does not require attribution modelling. It does not require arguing about which touchpoint deserves credit. It requires only measuring the delta between treatment and control.

The measurement can be rigorous. Random assignment ensures the groups are comparable. Statistical testing ensures the difference is real rather than noise. Holdouts can be maintained over time to verify that uplift persists rather than fading. The methodology is well-established. The disputes dissolve.

“You will need too much data. What about privacy?”

A credible system must be built with governance as a first principle, not an afterthought.

Context Graphs are designed for structured intelligence and restraint. The architecture enforces data minimisation: only the signals necessary for decisions, nothing extraneous. Every action is traceable through the Decision Trace Graph — what data was used, for what purpose, with what outcome. The system can explain why any intervention was made.

“Restraint is intelligence” is a design principle, not a marketing slogan. Meridian knows when not to act. It recognises when another message would erode trust rather than build it. It understands that attention is finite and abuse is cumulative. This restraint is only possible with governance infrastructure that tracks attention budgets and respects boundaries.

The mature intelligence system is not the one that maximises messages. It is the one that maximises relationship value while minimising intrusion.

“Outcome pricing sounds like agency performance fees. We have been burned.”

Many organisations have been burned by performance promises that delivered excuses instead of results. Agencies committed to outcomes and then explained why external factors prevented delivery. The creative was not approved in time. The budget was cut mid-campaign. The market shifted unexpectedly. The competitor launched a promotion.

Meridian is different in kind, not just degree.

It is a productised model, not a services engagement. The intelligence is systematic, operating continuously through autonomous agents rather than depending on which account manager happens to be assigned this quarter. Outcomes are tied to customer profit metrics — retention, repeat purchase, contribution margin — not vanity ROAS that can be gamed through channel shifting or measurement manipulation.

The model improves daily through continuous learning. Agencies improve annually through staff changes and strategy reviews, if they improve at all. The feedback loops are not comparable.

And critically, the accountability is real. If Meridian does not deliver uplift, it does not get paid. The excuses do not matter because the measurement is objective. Either the treatment group outperformed the control group or it did not. The results speak for themselves.

“What if the model fails?”

This objection contains its own answer.

If Meridian fails, the vendor does not get paid. The brand has risked nothing except the time spent on integration and the opportunity cost of the pilot period. The financial exposure sits entirely with the vendor.

This is precisely the point of Alpha pricing. It transfers risk from the brand to the provider. Traditional enterprise software puts the risk on the buyer: pay upfront, hope for results, absorb the loss if results do not materialise. Alpha pricing inverts this: the vendor invests first, delivers results, and gets paid only when value is created.

The willingness to offer Alpha pricing is itself a signal of confidence. Vendors who refuse outcome-based structures are implicitly admitting uncertainty about whether their products actually work. Vendors who accept are making a statement: we believe in this enough to bet on it.

“Why can only you build this? Will everyone else not copy it?”

Anyone can copy a feature. Very few can copy a signal moat.

Meridian’s edge comes from unifying three signal streams that most systems treat separately:

Engagement signals from owned channels — what customers open, ignore, click, dwell on, abandon, return to. These signals reveal attention patterns and preference dynamics that no other source can provide.
Channel signals from messaging infrastructure — responses to emails, behaviour on WhatsApp, interactions with push notifications, timing patterns across communication modes. These signals reveal channel preferences and responsiveness that engagement data alone cannot capture.
Discovery signals from on-site search and browse — queries entered, refinements made, products viewed, paths abandoned, recommendations clicked or ignored. These signals reveal intent and consideration that messaging data alone cannot see.

When these three signal streams feed the same Context Graph, learning compounds across the entire customer experience. The insight from a search query informs the email timing. The email response refines the product recommendations. The recommendation click updates the BrandTwin’s preference model. The cycle continues, each signal enriching the others.

This integration is the moat. Competitors who have only engagement data see part of the picture. Competitors who have only discovery data see a different part. The complete view requires all three — and building that integration from scratch takes years of infrastructure investment and data accumulation.

Every brand that joins the system makes the system smarter. Anonymised patterns across hundreds of brands create consumer intelligence that no single-brand solution can match. The model learns that customers who behave like this tend to drift for these reasons and respond to these interventions — not from one brand’s limited data but from the collective experience of an ecosystem.

This is precisely how adtech built its moat. Google and Meta aggregated signals across millions of advertisers to enable precision targeting that no single brand could achieve alone. The more advertisers participated, the smarter the system became. That network effect is why they dominate acquisition.

Meridian applies the same principle to retention. Every brand on the system makes the system smarter. Anonymised patterns across hundreds of brands create consumer intelligence that no single-brand solution can match.

This is the network effect that becomes defensible. Early adopters benefit from the intelligence that later adopters contribute. The gap compounds over time.

The Agentic Moment

Meridian is not a theoretical possibility for some distant future. It is becoming buildable now because of three forces that have converged simultaneously.

The first force is that agents can act.

For most of AI’s history, the technology was assistive. It could suggest, recommend, draft, and summarise. But action required a human in the loop — someone to approve the recommendation, send the message, make the decision. The human was the rate limiter. The AI was a productivity enhancer.

That boundary has shifted. AI agents can now execute workflows across systems. They can run experiments, adjust parameters, coordinate actions, and make decisions without requiring human approval at every step. The technology has moved from “chat” to “work.”

This matters because retention is not a one-time optimisation problem. It is a daily operating problem. Customers drift continuously. Attention decays continuously. The competitor’s offer arrives continuously. A system that requires human approval for every intervention cannot operate at the speed and scale that continuous drift demands.

Agents make continuous operation possible. They can process signals, update models, select actions, and execute interventions around the clock without human bottlenecks. They can run hundreds of micro-experiments simultaneously, learning from each result and adjusting the next action. They can maintain a living system rather than a series of campaign bursts.

Industry analysts predict that 15% of day-to-day work decisions will be made autonomously by AI agents within the next few years. Marketing — with its high volume of repetitive decisions and clear feedback signals — is the ideal proving ground for this shift.

The second force is that attention systems generate better signals.

Traditional marketing systems generate sparse signals. A customer either opens an email or does not. Clicks or does not. Converts or does not. The signal is binary and infrequent. Learning is slow because feedback is limited.

Systems like NeoMails and NeoBoost create frequent, lightweight interactions that produce rich signals. Not just whether the customer opened but how long they engaged. Not just whether they clicked but what they chose among alternatives. Not just whether they converted but what path they took to get there.

This signal density changes what is learnable. With sparse signals, models can detect only gross patterns. With dense signals, models can detect subtle dynamics — the early warnings of drift, the specific triggers that re-engage, the individual preferences that distinguish one customer from another.

More signal improves the Context Graph. Better context improves Meridian’s decisions. Better decisions improve outcomes. Better outcomes generate more engagement. The flywheel turns.

The third force is that Context Graphs can hold state and reason.

The breakthrough is not merely storing more data. It is storing meaning.

Decision traces capture why actions were taken, not just what actions occurred. Precedents accumulate so the system can learn from its own history. Confidence scores quantify uncertainty so the system knows what it knows versus what it guesses. Learning transfers across customers so insights from one relationship inform predictions about similar relationships.

This is what turns marketing from repeatable campaigns into compounding relationships. Without state, every interaction starts fresh. With state, every interaction builds on what came before. Without reason, the system cannot learn from its decisions. With reason, the system becomes wiser over time.

The protocols now exist to make this work at scale. Standards for agent communication, context sharing, and cross-system coordination have matured enough that building Meridian is an engineering challenge rather than a research problem.

Together, these three forces create the agentic moment: the point at which outcome-guaranteed marketing intelligence becomes technically and economically feasible. The components exist. The integration is buildable. The question is no longer whether this is possible but who will build it first.

And there is a compounding advantage for early movers. A well-built Meridian system gets better with every action — within a brand because decision traces reduce repetition and increase learning transfer, and across brands because anonymised patterns reveal behaviour dynamics no single brand can observe alone.

The early adopters will not just have a head start. They will have an expanding lead.

The Choice

Every CEO eventually asks the same question, even if they phrase it differently: “Why do we keep paying to reacquire customers we already had?”

The question usually surfaces after a board meeting where someone noticed that customer acquisition costs keep rising while retention rates stay flat. Or after a quarterly review where the marketing budget consumed its allocation with no visible impact on customer lifetime value. Or after a competitor’s move revealed how many “loyal” customers were actually just waiting for a better offer.

The question is simple. The answer has two parts.

Answer one: because you have been buying tools instead of outcomes.

The tools work. The platforms deliver messages. The journeys execute. The analytics report. But the accountability for whether any of it actually retains customers has remained with your internal team — a team constrained by bandwidth, turnover, campaign-mode operations, and cost-centre economics. The tools gave you capability. Capability is not the same as results.

Answer two: because there has not been an alternative.

The martech category was built to sell capability. No vendor offered outcome guarantees because the business model did not support them. Fixed SaaS fees do not create incentive alignment. Vendors who tried performance pricing could not make the economics work because their systems were not built to deliver reliable outcomes.

That is changing.

The choice is now real. Two paths exist.

The first path: buy platforms, deploy agents, train the team, execute brilliantly, build your own Alpha.

This is the right choice for organisations that want control over their marketing operations. They want internal capability that compounds over time. They want ownership of strategy, execution, and learning. They have the patience to build the expertise and the stability to retain it.

Agentic Marketing gives these organisations leverage. The tools are powerful. The AI amplification is substantial. Success is possible for those who can achieve sustained excellence in execution.

The second path: give a portion of your customer base, measure the delta, pay only on profit improvement, let Meridian and NEO deliver your Alpha.

This is the right choice for organisations that want outcomes without building internal capability. They want to convert marketing from cost centre to profit engine without undertaking multi-year transformation programmes. They want accountability transferred to someone willing to bet their revenue on results.

NeoMarketing gives these organisations certainty. Not hope. Not optimisation. Certainty that if results are not delivered, payment is not required.

Most organisations will use both paths.

Start with platforms and agents to build internal understanding. Learn what works. Develop marketing operations competence. Establish baselines.

Then graduate to NeoMarketing when you hit the ceiling of what internal teams can achieve — or when you simply want guaranteed outcomes rather than optimised effort. Use Meridian for Best customers when N=1 intelligence justifies the investment. Use NEO for Rest and Test customers where scalable recovery is the goal.

The paths are not mutually exclusive. They are complementary. The platform provides foundation. NeoMarketing provides guarantee. Together, they address the full spectrum of capability and accountability.

The unifying truth is that both paths serve the same mission: Never Lose Customers.

Agentic Marketing says: “We will give you the tools to pursue that mission.”

NeoMarketing says: “We will deliver that mission and prove it with measurement.”

The question is not which approach is correct. Both are correct for different situations. The question is: who do you want to be accountable?

If you want accountability to rest with your internal team, Agentic Marketing gives them the best possible tools and AI amplification.

If you want accountability to rest with a vendor who only gets paid on results, NeoMarketing provides that structure — and the intelligence to back it up.

Either way, the $500 billion AdWaste crisis is addressable. Either way, customer retention can become a profit engine rather than a cost line. Either way, the cycle of lose-and-reacquire can be broken.

The only question is who is on the hook for making it happen.

The Invitation

This essay has not described a product available for purchase today. It has described a direction — a vision of what marketing intelligence can become when accountability replaces activity as the organising principle.

The elements are falling into place. Context Graphs are deployable. BrandTwins are constructible at scale. Agents can act autonomously. Alpha pricing is commercially viable. The question is no longer “is this possible?” but “who will build it first, and who will adopt it early enough to capture the compounding advantage?”

The marketing technology category is about to split into two worlds.

The first world is utilities. Platforms priced on inputs. Capability commoditised over time. Feature parity and price pressure. Success depending entirely on perfect execution by the buyer. In this world, vendors compete on price because they cannot compete on outcomes. Margins compress. Differentiation disappears. The category becomes infrastructure — necessary, but undifferentiated.

The second world is outcomes engines. Intelligence priced on verified uplift. Built on proprietary models and compounding learning. Premium positioning justified by accountable delivery. Success depending on results, not activity. In this world, vendors who can guarantee outcomes command premium economics. Those who cannot become utilities.

Meridian is a bet on which world wins.

For CMOs:

Start demanding accountability from your vendors. Ask the uncomfortable question that most vendors hope you will not ask: “Will you take outcome pricing? Will you bet your revenue on our results?”

The vendors who refuse are communicating something. They are signalling their own confidence boundaries. They may have good products. They may have sophisticated features. But they are not willing to stake their economics on whether those products actually deliver what they promise.

The vendors who accept are making a different statement. They are expressing confidence born of capability. They are aligning their success with yours. They are signalling that they have something proprietary — something that justifies the exposure of outcome accountability.

When you find vendors willing to take that bet, you have found vendors worth serious consideration.

For CEOs and CFOs:

Marketing does not have to be a cost centre.

For decades, you have approved marketing budgets with limited visibility into what that spend actually produces. You have seen dashboards full of metrics that do not connect to the P&L. You have wondered whether all those martech subscriptions actually move the needle or merely create activity that looks like progress.

There is a different model. Customer retention is measurable. Profit improvement is attributable. Vendor economics can be aligned with your outcomes.

The question is not “what did we spend?” It is “what did we earn?” And with the right contract structure, that question becomes answerable.

The CFO who asks “what is the ROI of our martech stack?” typically receives vague answers about brand building and customer experience. The CFO who asks “what is the Alpha our retention system generated?” receives a number — verified uplift over control, measured in profit, defensible in any audit.

That is the difference between buying capability and buying outcomes.

For the industry:

The shift that is coming will not be comfortable for everyone.

Vendors whose products do not reliably deliver results will struggle with outcome pricing. They will resist the shift, arguing that attribution is impossible, that too many factors influence retention, that outcome-based models are impractical. These arguments reveal more about the vendor’s confidence than about the model’s feasibility.

Vendors whose products genuinely work will embrace outcome pricing because it differentiates them from competitors who cannot make the same offer. They will welcome the shift because it rewards capability over marketing, results over rhetoric.

The market will sort this out. Buyers who demand accountability will find vendors willing to provide it. Buyers who accept capability-without-accountability will continue to hope for results. Over time, the former will outperform the latter. The evidence will accumulate. The category will split.

The closing thought:

Meridian turns “Never Lose Customers” from a manifesto into a procurement category.

It is the answer to the CMO who is tired of tools that require perfect execution to deliver results. It is the answer to the CFO who is tired of marketing spend that disappears without accountability. It is the answer to the CEO who is tired of asking why customer acquisition costs keep rising while customer retention keeps stagnating.

The future of marketing is not better campaigns. It is not more sophisticated segmentation. It is not AI-generated content at scale.

The future of marketing is outcome accountability — the principle that vendors should share in the results they claim to produce, that intelligence should be measured by what it delivers rather than what it promises, that “Never Lose Customers” should be a contractual commitment rather than a conference keynote.

For those ready to stop buying tools and start buying outcomes, the architecture is emerging. The economics are proven.

Meridian is what that future looks like.

Never Lose Customers. Never Pay Twice. Never Pay Fixed.

Share this:

Like this: