The MaaS Manifesto

Why Made as a Service Will Systematically Replace Large Portions of SaaS

A whitepaper by Etienne de Bruin

Jan 6th, 2026

Executive Summary

Software as a Service solved a 1990s problem: how to distribute complex software when infrastructure, deployment, and maintenance were prohibitively expensive.

It succeeded by centralizing ownership and standardizing behavior.

For two decades, businesses accepted this trade: lower friction in exchange for compromise. They rented generic tools, adapted workflows to vendor assumptions, and accumulated operational overhead in the form of integrations, administrators, and compliance scaffolding.

That trade no longer reflects the underlying economics.

By 2026, advances in artificial intelligence have materially reduced the marginal cost of creating, modifying, and maintaining software. While human oversight remains essential, the mechanical labor that once dominated software economics has been partially automated. The consequence is not faster SaaS—it is a different acquisition model altogether.

I call this model MaaS: Made as a Service.

MaaS replaces software rental with software stewardship. Customers subscribe not to a static product, but to an ongoing capability: software continuously synthesized, governed, and evolved to match their exact workflows—deployed into infrastructure they control, with full ownership of data and logic.

This paper argues that MaaS is not a niche alternative. It is the natural successor to SaaS across any domain where:

• Workflows materially differ across organizations
• Competitive advantage lives in process
• Data context matters for AI systems
• Integration overhead dominates subscription cost

SaaS will not disappear. But its domain will narrow considerably.

The shift is not ideological. It is economical.

Part I: SaaS as a Scaling Technology—and Its Structural Limits

SaaS Solved Distribution, Not Fit

SaaS achieved scale by sharing code across customers. This enabled:

• Lower upfront cost
• Predictable pricing
• Centralized maintenance
• Rapid deployment

But shared codebases impose unavoidable constraints:

• Features must converge toward the median user
• Roadmaps reflect aggregate demand, not individual need
• Customization becomes configuration, not design
• Data models harden around vendor priorities

These constraints are not failures of execution. They are properties of the model.

The Hidden Cost Stack

SaaS pricing understates total cost by externalizing key burdens:

The result is an inverted cost curve: subscription fees shrink relative to the organizational cost of operating the tools.

In 2025, enterprises spent an estimated $3.80 on integration and middleware for every $1.00 spent on actual SaaS subscriptions. The connective tissue costs more than the organs.

This is why enterprises increasingly spend multiples of their SaaS fees on integration, analytics, and internal enablement.

The Concentric Circles of Dependency

The SaaS landscape has consolidated into concentric circles.

At the center sit platform giants: Salesforce, Microsoft, Google, Oracle, SAP. Their gravity pulls in smaller vendors who build integrations, extensions, and complementary tools. These second-tier companies depend on platform access, API stability, and ecosystem placement.

Around them orbit the integration layer: the middleware providers, the consultancies, the "implementation partners" who make their living connecting systems that don't want to be connected.

Customers orbit furthest from the center, subject to forces they don't control:

• Pricing changes ripple outward without negotiation
• Feature deprecation happens on vendor timelines
• Data portability remains theoretical
• Switching costs compound over time
• Security vulnerabilities in any ring affect all rings

This architecture concentrates value at the center while distributing costs—and risks—to the periphery.

The Data Jail Problem

In the AI age, data is fuel. But traditional SaaS models act as data jails—proprietary APIs and rigid schemas prevent AI agents from seeing the full context of a business.

Your customer data lives in Salesforce. Your project data lives in Asana. Your financial data lives in QuickBooks. Each system guards its silo.

SaaS has become a bottleneck to intelligence. The very platforms that promised to unlock your business now prevent your AI systems from understanding it.

Part II: The AI Shift—What Changed, Precisely

What AI Actually Changed (Without the Hype)

AI did not eliminate the need for engineering judgment. It did reduce the cost of execution.

Specifically:

• Translation from intent to code is faster
• Iteration cycles are shorter
• Large codebases are more legible to machines than to humans
• Routine refactoring and dependency management can be automated

This collapses two historic constraints:

1. Time-to-first-usable-system
2. Ongoing maintenance labor

Custom software is no longer defined by sunk cost.

Maintenance Is Now a Governed Process, Not a Tax

The historical objection—"custom software rots"—assumed maintenance was manual.

In the old model, custom software was a liability. It required human developers to patch vulnerabilities, update dependencies, fix "bit rot," and adapt to changing operating systems. A custom application built in 2020 might be unmaintainable by 2025—not because it stopped working, but because the humans who understood it moved on.

This objection has dissolved.

Modern MaaS systems combine:

• Continuous testing
• Automated dependency monitoring
• Security scanning
• AI-assisted refactoring
• Human architectural review

Maintenance shifts from reactive firefighting to continuous regeneration.

This does not eliminate risk. It isolates it.

The New Unit Economics

The math has inverted completely.

The Old Calculation:

Consider building a custom alternative to a typical SaaS tool:

Against a $300/month SaaS subscription ($3,600/year), the breakeven was 35 years. Custom development was economically irrational.

The New Calculation:

Where the 840 → 13 Comes From

This isn't a claim that AI codes 65x faster. It's three efficiencies compounding:

1. AI-assisted development: 3-5x faster
   Documented productivity gains from current tooling—Cursor, Claude, Copilot. Real, measurable, widely reported.
2. Scope reduction: 10-15x less code
   MaaS doesn't rebuild Salesforce. It builds exactly what one client needs—often 10-20% of a typical SaaS feature set. No admin console for 10,000 users. No enterprise SSO you'll never use. No features designed for someone else's workflow.
3. Reusable foundations: 2-3x savings
   Authentication, billing, compliance, audit logging, hosting—these are composed from existing primitives, not built from scratch.

The compound effect: 4x faster × 10x less scope × 2x reuse ≈ 80x

The old 840 hours assumed building a full-featured application from scratch, designed to serve many use cases, with custom infrastructure.

The new 13 hours reflects synthesizing a fitted application for one client's specific workflow, on top of shared foundations, with AI assistance.

The speedup isn't magic. It's compounding leverage across a fundamentally different approach.

The breakeven against that same SaaS subscription is now 14 months. And in year two and beyond, the MaaS solution costs less than the SaaS rental.

For a team of 50 using a specialized SaaS tool at $50/seat/month, the annual cost is $30,000. A MaaS-generated equivalent costs roughly $4,500 to synthesize and $2,400/year in autonomous maintenance. The ROI is realized in a single quarter.

Part III: MaaS Defined—Precisely and Defensively

MaaS Is Not "Custom Dev Rebranded"

MaaS differs from traditional bespoke development in five critical ways:

1. Shared Foundations Common logic (authentication, billing, compliance, audit trails) is reused across clients. The wheel is not reinvented.

2. Unique Orchestration Workflow logic and interfaces are custom, not templated. The surface is shaped to fit.

3. Continuous Stewardship Evolution is included, not renegotiated. The relationship persists.

4. Explicit Governance Testing, security, and compliance are automated by default. Quality is structural.

5. Customer Sovereignty Code, data, and deployment belong to the customer. Ownership is real.

MaaS achieves scale through learning accumulation, not feature accumulation.

How It Works

Intent-Based Discovery

MaaS engagements begin with understanding. What does the client actually need? Not what features do they want—what problems are they solving, what workflows are they enabling, what outcomes matter?

This phase is conversational, not bureaucratic. You describe the problem; the synthesis engine generates potential solutions. Clients can interact with working prototypes within hours.

Rapid Synthesis

AI-assisted development produces working software in timeframes that would have seemed impossible five years ago. A typical engagement delivers a functional application within days. The era of 18-month implementation projects is over.

Sovereign Deployment

Applications can run on customer cloud accounts, customer servers, customer security perimeters. Customers are not tenants in someone else's building; they own the property.

Continuous Evolution

The relationship continues. As needs change, the software adapts. Monthly fees cover not just maintenance but modification. The software grows with the business because the business and the maker remain connected.

The Three Tiers

Tier 1: Micro-Apps ($100-200/month) Single-purpose tools that do one thing perfectly. A custom pricing calculator for a specific manufacturing niche. A client portal tailored to exact service delivery. An inventory tracker that matches actual warehouse layout.

Tier 2: Integrated Workflows ($300-500/month) Multi-function applications that replace entire categories of SaaS spend. A custom CRM/Project Management hybrid built specifically for architectural firms. A complete operations platform tailored to restaurant group needs.

Tier 3: Enterprise Engines ($1,000-3,000+/month) Mission-critical systems requiring sophisticated logic, multiple integrations, and rigorous compliance. Industry-specific platforms with automated SOC2 compliance. Custom ERP modules that actually match how the business operates.

Part IV: The SaaS CEO Rebuttal (Steel-Manned)

"This sounds like old-school custom software with AI lipstick. SaaS won because customers don't want to own software—they want outcomes. Your model collapses under complexity and risk."

Response

This rebuttal assumes three things:

1. Custom software cannot scale safely
2. Customers prefer abstraction over control
3. Shared infrastructure is inherently more efficient

All three were true in 2015. None are categorically true now.

• MaaS does not ask customers to operate software—only to own it
• Governance is automated, not delegated to hero engineers
• Complexity exists today in SaaS—it is simply hidden inside integrations, admin roles, and brittle workflows

SaaS did not eliminate complexity. It displaced it.

MaaS internalizes that complexity where it can be systematically managed.

The question is not whether custom software can scale. The question is whether the hidden costs of SaaS—the integration tax, the workflow inversion, the data jails, the shared security exposure—remain acceptable when an alternative exists.

For an increasing number of organizations, they do not.

Part V: Hostile Q&A

Q1: "Isn't this just fragmenting the ecosystem into unmaintainable snowflakes?"

No.

Surface behavior is unique. Foundations are shared.

MaaS platforms reuse hardened primitives and compliance logic while allowing orchestration to vary. This is closer to how modern infrastructure works (Kubernetes + custom services) than artisanal development.

Q2: "What about security? Isn't bespoke code more vulnerable?"

Shared SaaS platforms create shared blast radius. Every supply chain attack on a major vendor reminds us: millions of customers exposed simultaneously.

MaaS isolates risk:

• Smaller attack surfaces
• No cross-tenant exposure
• Auditable code paths
• Explicit threat models

Security improves when responsibility is explicit, not abstracted.

Q3: "How does this scale profitably?"

SaaS scales through sameness. MaaS scales through reuse + automation.

The cost curve bends through:

• AI-assisted synthesis
• Reusable logic primitives
• Automated governance
• Declining marginal modification cost

The unit of scale is not customers—it is knowledge.

Q4: "Won't customers regret owning software?"

Only if ownership implies burden.

MaaS separates ownership from operation. Customers control destiny without bearing daily toil. They own the code and data; the MaaS provider handles the maintenance.

This is analogous to owning a home with a property management company versus renting an apartment. Ownership without operational burden.

Q5: "Why hasn't this worked before?"

Because three prerequisites were missing:

1. Cheap synthesis — AI collapsed the cost of code generation
2. Autonomous maintenance — Agentic systems handle ongoing updates
3. Governance infrastructure — Automated security and compliance tooling matured

All three now exist. The economic foundation has shifted.

Part VI: The Transition

Dismantling the Concentric Circles

MaaS doesn't just compete with SaaS—it dissolves the architecture that makes the concentric circles possible.

When custom software is affordable, the gravitational pull of platform giants weakens. Businesses no longer need to orbit Salesforce to access CRM functionality. They don't need to pay the integration tax to connect systems. They don't need to accept "good enough" from the SaaS establishment.

The circles collapse from the outside in.

Who Moves First

The Underserved SMBs Small businesses with "weird" workflows that never fit into Salesforce or Monday.com. They want software that does exactly what they need—nothing more, nothing less.

High-Alpha Professional Services Firms that want to codify their "secret sauce" into proprietary tools. Agencies and consultancies that understand custom solutions deepen client relationships and increase firm valuation.

Security-Conscious Enterprise Organizations tired of supply chain attacks on major SaaS vendors. Departments that want audited, isolated codebases under their own control.

Regulated Industries Healthcare, finance, and government—where compliance is easier to achieve with software designed for specific regulatory requirements than with generic platforms jury-rigged for regulated environments.

What SaaS Retains

SaaS will not disappear. But its domain contracts to areas where structural advantages remain:

Commodity Utilities Email, basic document editing, video conferencing. Functions that are genuinely universal with no competitive differentiation.

Network-Effect Platforms LinkedIn, Slack (for cross-company communication), marketplaces. The value is the people, not the code.

Heavy Compute Applications requiring massive proprietary infrastructure—video rendering, large-scale simulation, frontier AI training.

Everything else is vulnerable.

Market estimates suggest 40-60% of current SaaS revenue addresses segments where MaaS alternatives are viable. The transition will be measured in years, not months—but the direction is clear.

Conclusion: SaaS Was an Optimization—Not an End State

SaaS was the correct solution for its time.

But it optimized distribution, not fit. Scale, not sovereignty. Uniformity, not advantage.

The era of rental is giving way to the era of stewardship.

MaaS does not replace SaaS everywhere. It replaces it where compromise is expensive—where the bloat tax, the integration tax, the workflow inversion, the data jails, and the shared vulnerabilities cost more than businesses have been willing to admit.

The future of software is not owned by vendors. It is made, governed, and evolved in partnership.

We called the last era "democratization." It wasn't. It was standardization with good marketing.

The Sovereignty Era offers something better: software that is truly yours.

SaaS was software for everyone.

MaaS is software made for you.

The transition has begun.

About the Author

Etienne de Bruin is the founder of 7CTOs, a global community of technology leaders focused on growth, learning, and peer support. He is also co-founder of LevelsOS LLC, a technology leadership certification company, and creator of The CTO 1000, a valuation and ranking system for technical leaders.

With deep experience as a technical leader, executive coach, and community builder, Etienne works at the intersection of technology strategy and leadership development. His work focuses on helping CTOs and technical executives navigate the rapidly evolving landscape of software, AI, and organizational design.

He is the author of Liquid, an Amazon bestseller exploring consciousness and technology.

Etienne is based in Salt Lake City, Utah.

Contact: etienne@7ctos.com

© 2026 Etienne de Bruin. This whitepaper may be freely distributed with attribution.