Can Wavesteam operate and maintain a system after launch?
Yes, but first separate defect warranty, system maintenance, production operations, and feature development. Wavesteam can provide post-launch technical support only for the services, hours, and measurable targets stated in the agreement. “Free operations” does not define 24/7 staffing, cloud changes, recovery time, or third-party responsibility.
Warranty corrects reproducible errors within the delivered scope. Maintenance covers operating-system, middleware, certificate, and dependency upkeep. Production operations covers monitoring, incidents, backups, restoration, and capacity. A new report, workflow, or interface is product development. Each needs different people, response, and pricing. A contract saying only “one year of maintenance” leaves database scaling and supplier API changes disputed.
When planning milestones, resources, and acceptance, also compare Does Wavesteam continue supporting a system after it goes live? and Can response and restoration targets be written into the support contract?; the linked guidance adds context that should be considered in the same decision.
| Service | Typical work | Best fit | Define explicitly |
|---|---|---|---|
| Defect warranty | Correct reproducible code error against acceptance | Every new delivery | Period, defect definition, reproduction, exclusions |
| Business-hours maintenance | Inspection, patches, certificate alerts, logs, release support | Non-critical internal systems | Hours, notice, third-party cost |
| Production operations with SLA | Monitoring, on-call, incident levels, recovery, review | Revenue, transactions, equipment, production | SLI/SLO, response, restoration, consequence, exercises |
| Continuous iteration | Features, UX, performance change | Growing products | Capacity allowance, priority, separate acceptance |
The right level begins with the effect of one hour of downtime. An office administration tool may need only business-hours response. An order, energy, or production system running overnight requires separate investment in on-call staff, redundancy, and recovery rehearsal. “24/7” is a staffed and authorized operating capability, not four characters in a contract.
Turn “stable” into user-facing measures, such as monthly success for a core ordering flow, with stated exclusions for planned maintenance, client networks, and supplier outages. Google's SLI, SLO, and SLA guidance distinguishes a measure, objective, and contractual consequence. Record core-flow success, P95 latency, alert-to-response time, restoration, change failure, backup result, and tested recovery. DORA research is useful context for delivery and recovery performance, not a universal pass mark.
Keep production ownership clear. The client controls or can receive the cloud, domain, repository, and monitoring accounts; Wavesteam uses named least-privilege access. Approve and audit high-risk changes. A client-hosted environment needs a lawful, secure, revocable operations route. Wavesteam can diagnose and coordinate a failed messaging, mapping, or cloud supplier, but cannot promise the supplier's restoration time.
For a new project, launch readiness includes monitoring, contacts, backup, rollback, and known issues. Real alerts during the initial period reveal noise and blind spots before long-term service is agreed. For a system built elsewhere, Wavesteam first inventories code and assets, examines dependencies, restores a backup, and performs one controlled release. We do not promise a recovery target for an unobservable system with missing source or access.
At termination, hand back code, deployment guide, account and configuration register, runbook, unresolved issues, and recent backup, and revoke Wavesteam access. The client can self-operate or change provider without losing control. Specific service levels remain in the project agreement.