Does Wavesteam continue supporting a system after it goes live?
Yes, but launch support, defect warranty, production operations, and new development must be defined separately. Wavesteam does not disappear at handover and does not retain source, accounts, or data to force renewal. Before launch, both sides choose coverage, hours, and responsibility. The client can continue with Wavesteam, take over internally, or appoint another team.
Launch-window support covers deployment, observation, rollback readiness, and immediate issues. Warranty corrects reproducible deviations from the accepted scope. Operations covers monitoring, incidents, backup and recovery, capacity, and security maintenance. New workflows, reports, and platform adaptations are iterations. A vague “after-sales support” cannot decide certificate expiry, supplier API change, or scaling.
When planning milestones, resources, and acceptance, also compare Can Wavesteam operate and maintain a system after launch? and How long should software maintenance and production operations continue after delivery?; the linked guidance adds context that should be considered in the same decision.
| Continuation | Fit | Wavesteam | Client |
|---|---|---|---|
| On-demand after warranty | Non-critical, low-change system with internal monitoring | Estimate a ticket as defect or change | Monitor, back up, renew, triage |
| Business-hours maintenance | Internal or low-peak service | Inspection, alerts, release, agreed upkeep | Risk outside coverage and business decisions |
| SLO/SLA operations | Revenue, equipment, production | On-call, incident levels, recovery, review, exercises | Cost-risk choice, supplier contracts, authorization |
| Client or new-team takeover | Existing engineering and operations | Inventory, documentation, training, accompanied release, access removal | Continuing staff, process, production ownership |
There is no universal free one-to-three-month package. The contract states warranty start, end, defect, and exclusions. Warranty does not automatically include monitoring, cloud renewal, or night duty. A feature requested during warranty is not automatically free, and a true accepted-scope defect should not be relabelled as a paid addition.
The service schedule names contact channel, event severity, coverage, first response, mitigation or restoration, escalation, measurement, exclusions, and consequence. Google's SLI/SLO/SLA guidance explains why an SLA includes consequences, not only a technical target. The client provides operating hours, critical business, and interruption tolerance; Wavesteam derives user-facing measures and priced service options from monitoring and risk rather than asking the client to invent technical SLOs.
Initial monitoring covers core flows, errors, latency, queues, resources, backup, and certificates and verifies that alerts reach a person. Incident records retain impact, timeline, action, and restoration. Significant events receive a blameless review that improves detection and runbooks. DORA research can inform change and recovery observation without supplying a universal target.
Wavesteam can diagnose and degrade around payment, messaging, cloud, or store failure, but cannot guarantee the supplier's restoration. The client should hold those accounts and contracts for direct escalation.
From day one, the client owns or contractually controls repositories, cloud, domains, developer accounts, and business data, and Wavesteam uses named revocable roles. Handover includes source, build and deployment, architecture and data, secure configuration references, monitoring, recovery, suppliers, known issues, and runbooks. A receiving team proves readiness by building, deploying, rolling back, restoring a backup, and handling a simulated alert before Wavesteam access is removed.