Governance & Center of Excellence (CoE)
Governance & Center of Excellence (CoE)
As Datadog expands across an organization — more teams, more services, more products — the problems of ungoverned growth start to appear: inconsistent tags that make it impossible to filter by team or environment, dashboards nobody maintains, duplicate monitors nobody understands, and costs that can’t be allocated because nobody defined who owns what.
The observability Center of Excellence (CoE) is the answer to that challenge. Apiwan designs and implements the operating model that turns Datadog into a shared platform, with clear standards, defined ownership, and governance processes that scale with the organization — without unnecessary bureaucracy.
Who is this service for?
Ideal for organizations that:
- Have Datadog expanding across multiple teams and need to avoid chaos before it becomes irreversible.
- Are detecting inconsistencies in their configuration: non-standard tags, duplicate monitors, outdated dashboards, uneven instrumentation across teams.
- Want to implement showback or chargeback for the observability platform by team, area, or product — but lack the necessary data structure.
- Their platform or SRE team needs to define how Datadog should be used correctly and how to transfer that knowledge to development teams.
- Are going through operational maturity and want observability to be a strategic asset for the organization, not just another tool each team uses their own way.
What does the service include?
Instrumentation standards and tag taxonomy
Tagging is the backbone of governance in Datadog. Without a consistent taxonomy, it’s not possible to filter by team, allocate costs, correlate incidents, or build reliable multi-service views.
We define and implement the organization’s tagging strategy:
- Mandatory tags (env, service, team, version, region) and naming conventions
- Business tags for cost allocation (product, cost-center, application)
- Implementation guides by resource type (hosts, containers, APM services, logs, custom metrics)
- Automated tag validation in CI/CD pipelines to ensure compliance before deployment
Ownership model and RBAC
We define who’s responsible for what in Datadog: which team owns each service, dashboard, monitor, and log pipeline. We implement Datadog’s role-based access control (RBAC) model to ensure each team has access to what they need, and no more.
- Roles and permissions by team and by function (developer, on-call, viewer, admin)
- Datadog Teams as the base organizational unit
- Service Catalog configured as the source of truth for service ownership
Usage policies and configuration lifecycle
We establish the platform’s operating rules: how monitors are created, who can modify production dashboards, when obsolete configurations get archived, and how new teams are onboarded onto the platform.
- Monitor-creation guides and alerting conventions
- Periodic review and maintenance process for dashboards and monitors
- Onboarding workflow for new services and teams
- Offboarding process: how to retire services from the platform without leaving orphaned configurations
Showback and chargeback model
With tags and ownership correctly configured, we implement the dashboards and reports that let each area see how much its use of the observability platform and the associated cloud infrastructure is costing.
- Consumption dashboard by team: hosts, APM, logs, RUM, DBM
- Datadog cost allocation model by business area
- Automated monthly usage and cost report by team
Technical training and Datadog certifications
The CoE only works if teams understand the platform. We design training programs tailored to the organization’s different profiles:
- For development teams: APM instrumentation, structured logging, RUM, and custom metrics
- For operations/SRE teams: monitors, dashboards, SLOs, incident management
- For managers and decision-makers: reading dashboards, SLOs, cost reports, and business impact
- Datadog certification prep: support for teams looking to get certified on the platform
Periodic maturity reviews and ongoing coaching
The CoE isn’t a finite-time project — it’s an operating model that evolves with the organization. Apiwan provides ongoing support through:
- Quarterly CoE maturity reviews with scoring by area
- Office-hours sessions for the team’s technical questions
- Model updates in response to new Datadog capabilities or organizational changes
- Review of critical con
CoE maturity model
| Level | Characteristics |
|---|---|
| 1 — Initial | Ad-hoc implementation, without standards. Each team uses Datadog in its own way. |
| 2 — Repeatable | Basic tags defined, some shared dashboards. Documented practices but not systematically applied. |
| 3 — Defined | Formal tagging standards, RBAC implemented, team onboarding process, clear service ownership. |
| 4 — Managed | Active showback/chargeback, observability quality metrics, periodic configuration reviews. |
| 5 — Optimized | Observability as a strategic asset: autonomous teams, documented continuous improvement, CoE as an internal benchmark. |
Apiwan assesses the organization’s current level during the initial Assessment and designs the roadmap toward the target level.
