Total cost model
The six cost centers in a realistic budget
Estimate a low, expected and high scenario for each center. Use product exposure and consequence—not excitement about the prototype—to choose the contingency.
Discovery and content
InitialUser journey, page purpose, real content and acceptance criteria
RecurringResearch, copy updates and content governance
Cost triggerGeneric output creates rework or fails to answer customer objections
Build and integration
InitialInterface, data, authentication, vendors and deployment
RecurringFeature changes, dependency updates and integration drift
Cost triggerThe prototype crosses into accounts, transactions or complex state
Quality and specialist review
InitialJourney, accessibility, security, performance and privacy assessment
RecurringRegression testing and review after material change
Cost triggerRisk grows faster than the original builder’s expertise
Infrastructure and vendors
InitialDomains, environments, storage, monitoring and account setup
RecurringUsage, seats, bandwidth, email, databases and observability
Cost triggerFree allowances end or product behavior creates expensive requests
Operations and support
InitialRunbooks, alerts, backups, recovery and support setup
RecurringIncidents, customer support and routine maintenance
Cost triggerReal users expect response and data durability
Change or exit
InitialPortable data, source access, licenses and migration path
RecurringKeeping exports, tests and documentation current
Cost triggerVendor limits, pricing or product needs change
A useful formula
Compare total ownership, not day-one invoices
A practical estimate is: creation plus verification plus twelve months of tools and operations plus an explicit change contingency. Add an exit allowance when the platform controls data, deployment or components that cannot be reproduced elsewhere.
Do not assume a more expensive initial build is safer, or a cheap build is automatically wasteful. The economic advantage appears when the system remains understandable, changes stay small, failures are recoverable and the product reaches evidence of demand sooner.
- Estimate the same scope for every delivery option
- Separate fixed costs from usage-sensitive costs
- Name who absorbs defects and maintenance
- Write the assumptions beside the number
Decision matrix
Budget differently by product exposure
The interface may look similar while the operating cost changes substantially. These are planning directions, not price bands.
| Product type | Cost emphasis | Commonly forgotten | Budget signal |
|---|---|---|---|
| Static campaign page | Content, design, accessibility and analytics | Domain ownership, forms and consent | Short lifespan with a clear archive plan |
| Company marketing site | Content system, SEO, performance and maintenance | Editing workflow and dependency updates | Named owner after launch |
| Authenticated SaaS pilot | Data, authorization, monitoring and support | Negative-path tests and recovery | Controlled user and data scope |
| Payments or marketplace | Transactions, fraud, roles and reconciliation | Idempotency and dispute operations | Specialist review and incident capacity |
| Sensitive-data product | Privacy, security, retention and governance | Processors, deletion and breach response | Domain-qualified advice before exposure |
Applied example
Why the cheapest prototype can become the expensive option
Two proposals deliver the same visible portal. One includes organization-owned accounts, tests, documentation and a client-operated handoff. The other includes only a live URL and promises fast future edits.
- The second proposal may have a lower initial invoice but leaves ownership and maintenance undefined.
- Future changes depend on one builder rediscovering undocumented behavior.
- Migration or incident recovery becomes an unpriced emergency rather than planned work.
- The correct comparison adds the missing verification, ownership and operating obligations to both scopes.
Plain answers
Cost questions
Is vibe coding cheaper than hiring a developer?+
It can reduce time for some discovery and implementation work. Total cost still depends on complexity, review, operations, ownership and the cost of errors; the categories are not interchangeable.
Can I stay on free plans indefinitely?+
Do not build a budget around that assumption. Model the usage level that would indicate product success and review current vendor terms directly.
What is the most important contingency?+
For a low-risk site it may be content and maintenance; for an application it is often data, security, recovery or migration. Tie contingency to consequence.