IT Operational Maintenance (MCO)

• Updates, security patches and version upgrades
• Tested backups and disaster recovery (DRP/BCP)
• 24/7 monitoring, alerting and incident response
• Documentation and reversibility: your system stays yours

Operational upkeep is often reduced to patching. In practice a serious contract covers four families of work with different rhythms and risk.

Corrective. Security patches for the operating system, database engines and application dependencies. The most visible and most regular part.

Preventive. Watching the thresholds that announce a failure before it happens: disk saturation, memory drift, certificate expiry, provider quota reached, a backup failing silently.

Evolutive. Major version upgrades, end of support for a PHP or Java version or a managed engine, replacing a component the cloud provider has deprecated. This is the part internal teams postpone the most, and the one that produces technical debt.

Curative. Detecting, qualifying and resolving incidents, with a commitment on response and clear reporting.

The three terms overlap and are often used interchangeably in tenders. The useful distinction is what the contract is about.

MCO covers infrastructure and technical components: keeping the platform running. Application maintenance covers business code: fixing and evolving the application itself. Managed services is the umbrella term for delegated operations as a whole, MCO included.

One thing to check before signing: the boundary between the two scopes. An application incident caused by an infrastructure upgrade falls on both sides of the line, and that is exactly where poorly written contracts produce back and forth while production is down.

An MCO fee is priced on scope, not on a ticket count. The variables that matter are the number of environments to maintain, how critical production is, the expected coverage hours, and the number of distinct technical components to track.

We price on quotation, after scoping the real perimeter. Per-ticket pricing has a structural flaw: it aligns the provider's interest with the number of incidents, not with their disappearance. A flat fee aligns both.

FinOps is included in our managed services contracts. Tracking the cloud bill is part of operational upkeep: a platform whose cost drifts unnoticed is not a platform under control.

Most operational upkeep work is repetitive and known in advance. Our environments are described as code and provisioned by our BAM platform, which has two direct consequences.

First, recovery after an incident is a rebuild, not an improvisation: the target environment is described, therefore reproducible. Second, common failures are handled automatically and a human steps in only when a decision is needed. That is what makes 24/7 coverage possible without billing a full on-call team.

On non-production environments, shutting them down outside working hours typically saves 50 to 80% on those environments.

We operate your platforms worldwide, in every cloud region, with no geographic limit.