An AEM site in production is never finished. With AEM as a Cloud Service the platform updates continuously, dependencies evolve, browsers change, business requirements shift and editors ask for new components. A project that ends at go-live leaves behind a system that still needs maintenance, and the cost of not doing it shows up as technical debt, risky releases and knowledge concentrated in a few people.
Our AEM managed services are an ongoing service, organized as a retainer, in which AEM developers with experience on Sites and Assets work in the client's repository following a shared backlog. They handle fixes, enhancements, dependency updates, pipelines, Dispatcher and editor support. The model is designed for those who do not want, or cannot, maintain a full AEM team internally.
The service works because it is explicit: scope, priorities, communication channels, release methods and service levels are defined at the start and reviewed periodically. Commercial terms, covered hours and response times are agreed case by case, and no generic values are promised outside a contract.
What AEM managed services include
The typical scope includes fixing defects, building new components and templates, evolving content models, maintaining OSGi bundles, Sling Models and clientlibs, managing Cloud Manager pipelines and changing the Dispatcher configuration. It includes support for editors on authoring problems and on the correct use of templates, Experience Fragments and Content Fragments.
The service may also include periodic preventive activities: reviewing Oak indexes, analyzing logs and slow queries, checking failed workflows, verifying cache hits and renewing certificates and integration credentials. These activities are not very visible, but they are the ones that prevent incidents, and they are planned in the backlog like any other item.
Continuous updates and release process
On AEM as a Cloud Service platform updates are continuous and managed by Adobe. The team's task is to keep the code compatible: periodically updating the SDK dependency, reviewing deprecated APIs, running regression tests and watching the outcomes of Cloud Manager pipelines. Letting updates pile up for months turns routine maintenance into a recovery project.
The release process is made repeatable: an agreed branching strategy, code review, automated tests in the quality gate, deployment to development and stage, functional verification and then production. Where the project allows, we reduce manual releases and dedicated windows, so that small changes reach production often and with limited risk. Program-specific details should be verified against Cloud Manager documentation [VERIFICARE].
Retainer model and shared backlog
In the retainer model the client has an agreed capacity, expressed in days or dedicated developers, to spend on a backlog that both sides can see. Priorities are decided by the client, and we contribute estimates, risks and sequencing proposals. A periodic alignment meeting reviews the backlog, releases and open issues, without turning into an empty ritual.
A retainer fits when the flow of requests is continuous and predictable. For more concentrated needs, such as a new site or a complex feature, a fixed-scope project makes more sense. The two models often coexist: a project for major evolution and a retainer for routine maintenance, with a single repository and a single pipeline.
Monitoring, incidents and on-call
Useful support includes the ability to understand what is happening in production. We set up consultable logs and metrics, where possible forwarded to the client's observability tools, alerts on errors and response times, and a clear procedure for reporting an incident. After every significant incident we produce a short root cause analysis, with corrective actions added to the backlog.
Out-of-hours on-call, if included, must be defined precisely: covered time bands, reporting channels, response times and the scope of what can be fixed without an ordinary release. We prefer a service with explicit boundaries to a promise of unlimited availability that no real contract supports.
Knowledge transfer and exit
A support provider should not become a point of dependency. For this reason the service includes up-to-date technical documentation, recorded architectural decisions, operating instructions for releases and incidents, and knowledge transfer sessions to internal teams. The repository and pipelines remain under the client's control, and every change goes through its review rules.
If in the future the client decides to bring the work in-house, change provider or reduce the scope, the transition must be possible without risk. We define a realistic exit plan from the start and revisit it when the context changes. This approach reduces initial distrust and, in our experience, makes the collaboration healthier over time.
FAQ
AEM managed services FAQ
It depends on required capacity, covered hours and service levels. We do not publish generic rates: after an initial analysis of the repository and the flow of requests we propose a scope with explicit terms.
Yes. They work in your repository, follow your branching and code review conventions and release through your Cloud Manager pipelines. This keeps control in your hands and simplifies any later transition.
It can be included, with time bands, channels and response times defined by contract. We do not promise generic availability: we specify what is covered and what requires an ordinary release.
The platform updates continuously. We keep the code compatible: periodic SDK updates, review of deprecated APIs, regression tests and pipeline checks. Updates are not allowed to pile up.
Yes, and it is a common path. After a migration or rebuild a stabilization and maintenance retainer can be activated, with the documentation and pipelines already prepared during the project.
Set up AEM managed services
Describe your flow of requests, environments and critical activities. We will propose a service model with explicit scope and levels.
Request AEM support