Datadog: project + rebuild ROI.
Hosts, custom metrics, log volume, APM traces, RUM, synthetics — they all bill separately. Fourteen probing questions. Output: 24-month Datadog projection, comparison to Prometheus + Grafana + Tempo + OpenObserve rebuild, payback month, 3-yr reclaim. The full memo includes architecture sketch + migration playbook + cardinality control patterns.
How this is calculated
Datadog list pricing per product line (infra, APM, logs, RUM, synthetics, DBM, cloud cost). Custom metrics surcharge applied beyond base allowance. 24-month projection assumes typical 25%/yr growth in monitored stack. Rebuild cost: Prometheus + Grafana + Tempo + OpenObserve productized stack on AWS, with 1-2 platform FTE depending on scale.
What this does NOT estimate
- Migration disruption to incident-response workflows
- Custom dashboard rebuild effort (typically modest with Grafana provisioning)
- Specific compliance overlays beyond stated
FAQ
When does rebuilding from Datadog make sense?
Above ~$300K/yr Datadog spend with growing stack, rebuild typically pays back in 8-14 months. Below $150K/yr, stay on Datadog — operational simplicity wins.
What about cardinality?
Custom metrics surcharge is the silent cost driver — most orgs find 30-50% of metrics are unused. Memo includes the cardinality audit playbook.
What's in the memo?
Cost projection, rebuild architecture sketch, migration playbook, cardinality control patterns, on-call workflow continuity.