Migrating from Datadog to Dynatrace: A Step-by-Step Guide

· By IMRAN HOSSAIN

Quick answer: You can migrate from Datadog to Dynatrace safely by running both platforms in parallel, discovering and mapping every metric, log and trace, automating configuration with APIs and Terraform, and using a staged cutover with rollback plans and data export for history preservation. Start with discovery, then map, validate, automate and cut over.

  • Key takeaways: run parallel, map telemetry exactly, automate with API/Terraform, keep a rollback window, and budget for conversion effort and data egress.
  • Use staged validation: smoke tests, synthetic checks, and SLO comparisons before final cutover.
  • Preserve historical data by exporting Datadog archives or forwarding to shared object storage during migration.

Why should you migrate from Datadog to Dynatrace?

Answer: You migrate when you need automated topology, Davis AI root cause, or unified full stack automation over manual instrumentation. Dynatrace emphasizes automatic discovery with Dynatrace OneAgent and Davis AI for causal analysis, reducing manual tagging and service mapping work. Datadog remains strong in dashboard flexibility, log pipelines, and third party integrations. Choose migration when your pain points are slow incident triage, incomplete service topology, or you want a single-agent approach to auto-instrumentation for hosts, containers and serverless functions. Migrating carries effort: you must map Datadog monitors and dashboards to Dynatrace metrics, reconfigure alerts, and plan for data retention differences. See the detailed comparison in the Datadog vs Dynatrace guide if you are still deciding between the platforms.

What discovery steps must you run before migrating?

Answer: Run a full inventory of hosts, containers, services, dashboards, monitors, traces, and custom metrics. Use agent inventories, API calls and tagging reports to list Datadog hosts, APM services, log sources, custom metrics and dashboards. Export lists via the Datadog API, identify owners for each dashboard and alert, and tag resources for migration priority. Include cloud integrations such as AWS CloudWatch metrics, Kubernetes clusters, and custom exporters. Run dependency mapping with traces and topology to understand service boundaries and ingress points. Document retention policies for logs and metrics and estimate egress if you export to S3 or GCS. If you need checklist examples for audit and governance steps, see the NetSuite Sage checklist for a migration-style audit approach you can adapt to telemetry.

How do you map metrics, logs and traces between Datadog and Dynatrace?

Answer: Map by semantic meaning, not exact metric names; map logs and traces to OpenTelemetry concepts where possible. Start by grouping Datadog custom metrics by purpose: system, application, business. For metrics, translate Datadog metric name, tags, units and aggregation to Dynatrace metric or custom metric equivalents, recording cardinality differences. For logs, map pipelines, processors and parsing rules; use the Datadog Logs API to export structured logs and re-ingest into Dynatrace via the Dynatrace Log ingest API or via an intermediary like Fluentd. For traces, prefer OpenTelemetry if you can, migrating Datadog APM spans to OpenTelemetry format using exporters or collectors; otherwise keep Datadog traces during the parallel run and validate span IDs and parent relationships before cutover. For practical examples of reformatting pipelines, see the guidance in the Sentry and Datadog integration notes which explain mapping event schemas.

How do you run a parallel deployment safely and for how long?

Answer: Run Datadog and Dynatrace in parallel until validation shows parity across SLOs and alerts, typically 2 to 12 weeks depending on scale. Deploy Dynatrace OneAgent incrementally across environments, mirror Datadog dashboards with equivalent Dynatrace charts, and forward logs and traces to both platforms. Validate by running synthetic tests and comparing SLI values, error rates, and latency percentiles. Maintain Datadog for incident triage and use Dynatrace for deep causal analysis during coexistence. Roll forward automated alerting gradually: start with notifications only to an audit channel before enabling production on-call. If you need a staged governance checklist for parallel runs, adapt steps from our HubSpot Pardot checklist for handoffs and owner signoffs.

How should you migrate dashboards and alerts?

Answer: Rebuild critical dashboards first, then secondary ones, and convert alerts into grouped problem notifications. Inventory dashboards, mark critical views and owners, and recreate dashboards in Dynatrace using the same SLIs, time windows and aggregation. For alerts, translate Datadog monitors into Dynatrace problem detection and Synthetic monitors. Use automation where possible, exporting Datadog dashboard JSON and translating panels to Dynatrace API requests, or rebuild with Terraform. Keep a canonical dashboard mapping file that records original metric names, tags and equivalent Dynatrace metric IDs. If you have custom APM or Sentry dashboards, see our guide on Transition from Sentry and the Frontend monitoring guide for converting user-experience views.

What automation scripts and tools should you use for migration?

Answer: Use API-driven automation with Terraform, GitHub Actions and vendor APIs to continuously apply and validate configuration. Use Terraform providers for Dynatrace and Datadog where available to export and import dashboard and alert definitions. Use the OpenTelemetry Collector to normalize traces and metrics during forwarding. Automate validation with GitHub Actions that run smoke synthetic checks and SLI comparisons and alert on drift. For log reingestion use Fluentd or Logstash to transform Datadog archive formats to Dynatrace ingest formats. Store your configuration as code in Git repositories and gate changes with PR reviews. For credentials and secrets, plan rotations and refer to migration guides like LastPass to 1Password and migrate to Bitwarden for securely transferring secrets during automation.

What hidden costs and trade-offs should you plan for during migration?

Answer: Hidden costs include data egress, duplicated ingestion during parallel runs, engineering time to rebuild dashboards, and retraining teams. Consider data egress if you export Datadog archives to S3 or GCS, and double ingestion costs while forwarding telemetry to both providers. Plan for time spent remapping high-cardinality custom metrics: excessive cardinality can dramatically increase bills in both Datadog and Dynatrace. A practical decision framework: 1) classify telemetry as critical, nice-to-have, or disposable; 2) keep critical telemetry during cutover; 3) drop or aggregate disposable streams. Also remember company age and vendor maturity can matter; Datadog was founded in 2010, and Dynatrace traces its origins to 2005, according to their Wikipedia pages, which shows long vendor histories and ecosystem maturity. Use those facts to assess vendor roadmaps and support lifecycles when negotiating contracts.

What is a reliable cutover checklist for the final switch?

Answer: Cut over only after successful parallel validation, a confirmed rollback window, and owner signoffs. Your cutover checklist should include: disable new instrumentation to Datadog or switch new hosts to Dynatrace OneAgent, update DNS or routing for synthetic checks, migrate alert notifications to on-call channels, freeze dashboard edits for 24 to 72 hours, validate SLOs and run final smoke tests, and archive Datadog dashboards and logs. Keep a communication plan with stakeholders and a defined rollback trigger such as significant SLO deviation or increased alert noise. For audit and governance style signoffs, adapt the sequence from the Infusionsoft to Pipedrive transition checklist, which shows how to manage owner acceptance and rollback approvals during migrations.

How do you prepare a rollback plan and when should you execute it?

Answer: A rollback plan must be automated, reversible, and time-limited. Maintain a scripted way to re-enable Datadog collection for hosts, restore Datadog alert routing and reassign on-call, and re-open dashboards for editing. Keep a snapshot of pre-cutover configuration and an automated script to reapply it via Datadog API or Terraform. Define objective rollback triggers like SLO breaches exceeding X percent over baseline or critical alerts that block operations. If a rollback is executed, keep Dynatrace in read-only mode to collect forensic data without affecting production alerts, and schedule a postmortem to identify migration gaps. For configuration snapshot techniques, reuse methods from the Sage Intacct migration approach to preserve state and enable fast restoration.

Which migration profile fits you, and what are simple decision rules?

Answer: Choose an approach based on team size, telemetry volume, and business criticality. Solo or small teams with under 50 hosts: do a phased migration, rebuild only critical dashboards and keep Datadog as archive. Mid-market teams with 50 to 500 hosts: run a 4 to 8 week parallel deployment, automate dashboards and alerts, and budget for an SRE to map custom metrics. Large enterprises with 500+ hosts and regulated data: plan a 3 to 6 month migration with cross-team owners, compliance checks and professional services. Decision rules: if you need automated topology and AI-assisted root cause, choose Dynatrace; if you need extreme dashboard flexibility and many third party integrations, stay with Datadog. Who should NOT use this migration playbook: organizations with trivial monitoring needs or fewer than 5 hosts where migration cost exceeds benefit.

Worked examples

Example: Small SaaS startup. You run 25 hosts and 5 Kubernetes clusters. A phased migration focusing on APM and critical dashboards, using Terraform to recreate 10 dashboards, will likely take 3 to 6 weeks and can be completed by a single SRE with part-time help from engineers.

Example: Mid-market ecommerce. You run 180 hosts, 40 custom metrics and Splunk for logs. Expect 8 to 12 weeks to migrate, with automation for dashboards and Fluentd for log forwarding, plus one month of parallel billing for duplicated ingestion.

What common mistakes do teams make during this migration?

Answer: Common mistakes include underestimating metadata and cardinality, not preserving log schemas, skipping parallel validation, rebuilding too many dashboards at once, and neglecting cost impact of duplicate ingestion. Teams often forget to capture tag schemas and owner contacts before deleting or reprocessing metrics. Another frequent error is enabling production alerts in the new system without first directing them to a test channel, which causes alert storms. Also, forgetting regulatory requirements for log retention and data residency can break compliance. Avoid these by freezing editing, tagging owners, exporting logs to an object store for archival, and validating SLO parity before full cutover. For dashboard migration patterns, review conversion examples such as those in our Datadog APM vs Sentry notes to learn how to prioritize views and reduce duplication.

Comparison table: Datadog vs Dynatrace for migration planning

Product Price range Best for Key strengths Main limits
Datadog Vendor pricing tiers, see vendor pages Flexible dashboards, integrations Rich integrations, powerful log processing, dashboarding Requires manual topology building at scale, cost with high cardinality
Dynatrace Vendor pricing tiers, see vendor pages Automated topology and AI analysis Automatic discovery with OneAgent, Davis AI causal analysis Less dashboard flexibility for ad hoc visualizations, different pricing model

What final validation and handover steps should you run?

Answer: Validate SLO parity, run final synthetic tests, and collect owner signoffs before decommissioning Datadog. Confirm that all critical SLIs match within an acceptable delta, perform load tests if relevant, and run incident drills to ensure alerting works. Freeze Datadog updates, retain an archive of dashboards and logs in object storage, and schedule a 30 to 90 day monitoring overlap depending on business risk. Document runbooks, update runbook owners, and train on Dynatrace Davis AI workflows. For configuration governance and handoff patterns, adapt the owner acceptance steps from the Sentry to Datadog migration guide to ensure clear responsibility transfer.

FAQ

Answer: Common voice-search questions and short answers are below, each starts with a direct answer and expands briefly.

  • Is Dynatrace better than Datadog? Short answer: It depends on your priorities. Dynatrace is better when you need automatic full-stack topology and AI causal analysis, Datadog is better for flexible dashboards and broad third-party integrations.
  • Who is Datadog’s biggest competitor? Short answer: Dynatrace is a primary competitor, alongside New Relic and Splunk Observability, depending on area of focus such as APM, logs or metrics.
  • Can I export logs from Datadog? Short answer: Yes, export via the Datadog Logs API or archive to S3/GCS, and then re-ingest or store for compliance and historical needs.
  • Who are Dynatrace’s biggest competitors? Short answer: Competitors include Datadog, New Relic, AppDynamics and Splunk Observability, with differences in automation and pricing models.
  • How long does a migration take? Short answer: It varies; small environments may finish in weeks, large enterprise migrations take months because of discovery, parallel runs and validation.
  • Will I lose historical data? Short answer: Not necessarily, but you must export or archive Datadog data before decommissioning, since historical data is not automatically ported.

Summary: Migrating from Datadog to Dynatrace is a multi-phase process of discovery, mapping, parallel validation, automation, cutover and rollback planning. Use agent-based OneAgent deployment, OpenTelemetry for traces, automated API/Terraform scripts, and a conservative parallel window to reduce risk. Preserve history by exporting Datadog data and keep a clear rollback plan during the final cutover. For a focused comparison of AI observability capabilities, see descriptive anchor text.

External references: For vendor histories, see the Datadog Wikipedia page and the Dynatrace Wikipedia page for founding years and company context.

Checklist:

  • Inventory hosts, services, dashboards and owners
  • Export Datadog dashboards, monitors, metrics and logs
  • Deploy Dynatrace OneAgent to a pilot group
  • Forward logs/traces to both platforms and validate SLOs
  • Automate dashboard and alert creation with Terraform and APIs
  • Run parallel for a defined period, then cut over with rollback window
  • Archive Datadog configuration and data after decommission

Internal resources you may need during migration include migration checklists and transition guides for related systems, such as the NetSuite Sage checklist, HubSpot Pardot checklist, Infusionsoft to Pipedrive, Sentry and Datadog, Sage Intacct migration, Sentry to Datadog, Sentry vs Datadog, LastPass to 1Password, migrate to Bitwarden, OneNote to Notion, UptimeRobot alternatives, Datadog APM vs Sentry, Frontend monitoring guide, and QuickBooks to Xero for related migration patterns.

Edit Template