Monolithic or modular architecture for growing web systems
The appropriate web architecture depends on change thresholds, teams, and operational maturity. Modularity is only worthwhile with independent responsibilities.
For management and product owners, the key factors in choosing between a monolithic or modular web architecture are "independent change cycle" and "controlled data sovereignty." "Premature extraction" serves as a counter-test.
Published: 3 min read · Author: Sebastian Geier
When should a growing web system remain monolithic and when should it become modular?
A growing web system should remain monolithic as long as changes can be delivered together and internal module boundaries sufficiently limit coupling. Standalone modules or services are worthwhile when change cycles differ, data ownership is clear, and operational isolation is necessary. Decomposition follows observed coupling, not a pre-selected architectural style.
Diagnostic Case: "Premature Extraction"
The catalog, shopping cart, and editorial content reside within a single application but are separated by clear internal interfaces. Extraction is only considered when the catalog needs to scale independently and be frequently published by a separate team. The shopping cart remains in the monolith because separation would currently only generate distributed transactions.
Premature Extraction
Premature Extraction A still-changing business boundary is fixed as a network contract, slowing down learning.
Distributed Monolith Separate deployments remain coupled via synchronous calls and shared data, requiring coordination of each release.
Operational Multiplication – Each new entity requires its own telemetry, access control, and recovery, without providing corresponding benefits.
Controlled Data Sovereignty
Evaluate change history and identify areas that are frequently modified together or repeatedly independently.
Stabilize the strongest candidate boundary internally via interfaces and separate tests within the monolith.
Extract a unit only if its benefits are proven, and then re-measure coupling, release flow, and operational effort.
Independent change cycle
Independent change cycle – The candidate is regularly modified independently and demonstrably slowed down by shared releases.
Controlled Data Sovereignty – The module can maintain its data and consistency rules without forcing continuous distributed write operations.
Operating costs borne – Deployment, monitoring, security, and fault diagnosis of the additional unit are covered by personnel and technology.
Operating costs borne
Control signal
Signal 1
Proportion of changes that unintentionally affect multiple functional modules and joint acceptances.
Control signal
Signal 2
Frequency of independent releases of an extracted unit without coordinated adjustments to adjacent components.
What to consider when discussing "Monolithic or Modular Web Architecture"
Maintaining a robust decision file for digital systems answers the next practical question: What information belongs in a robust decision-making document for digital systems?
Making performance budgets binding for new features continues this line of thought with another question: How do performance budgets for new website features become truly binding?
If you want to practically implement "Monolithic or Modular Web Architecture," you can refer to Robust Website Systems which focuses on "Architectural Boundaries and Scaling" and "Independent Change Cycle."
Conclusion: Monolith or Modular Web Architecture
A structured monolith is a viable architecture as long as its boundaries effectively manage changes. Distribution is a targeted response to demonstrated independence and not a maturity indicator.
Sources and Further Information
The primary sources define the technical framework for "monolith or modular web architecture."
OpenAPI SpecificationPrimary specification for machine-readable HTTP API contracts, including operations, data models, and error responses.
Choosing Technology: An Introduction – GOV.UK Service ManualOfficial guidance on prototyping integrations, carefully cutting components, and evolving via open standards.
14. Operate a reliable service – GOV.UK Service ManualOfficial standard for the operation, availability, recovery, and continuous improvement of reliable services.
Key Thesis
A well-structured monolith is often the more advantageous starting point. Modules pay off when parts can be modified, operated, or managed independently by separate teams.
What This Is Not About
The choice is not a competition between an outdated monolith and a modern modular world. The number of functions or developers alone does not justify system decomposition.
What it's about
Change dependencies and the costs of independent delivery are relevant. Modularity is useful when stable functional boundaries support their own lifecycles and the additional operation is justified.
More insights
Platform strategy & build vs. buy
When does integration become more expensive than developing from scratch?
A separate evaluation step for "Monolithic or Modular Web Architecture" is: At what point does integration become less economically viable than developing from scratch?
Platform strategy & build vs. buy
Multi-Tenancy from the Start or Only When Needed?
"Monolithic or Modular Web Architecture" is supplemented by a separate decision: Should multi-tenancy be built immediately or added only when specifically needed?
Insights Overview
All VELUNO Insights at a Glance
Further analyses on Website Systems, digital visibility, and robust working models.
Operating costs incurred: The path to testing
Before decomposing a system, real change couplings and operational obligations should be considered together. An architecture cross-section review can show which boundary is sufficient internally and which would justify a true extraction.