Pricing analysis, live-verified September 2026
Datadog Pricing in 2026: What You Actually Pay Per Host
We modeled the real bill at three team sizes and checked it against three rivals and 1,500+ aggregated customer reviews.
Summary
Datadog pricing looks simple at $15/host/month until APM ($31/host/month), log indexing ($0.10/GB), and custom metrics stack on top. At 40 hosts with APM, list price runs $1,840/month before a single log line ships. We modeled the bill at 3 team sizes, checked it against 884 G2 reviews and 354 TrustRadius reviews, and priced out New Relic, Grafana Cloud, and Dynatrace as the honest alternatives.
Datadog pricing starts at $15/host/month for Infrastructure Pro, but that number is a teaser. Add APM ($31/host/month), log indexing ($0.10/GB), and custom metrics, and a 40-host team lands near $46/host/month before logs. Under 50 hosts it's fast and fair. Past 200 hosts, the per-host-plus-per-signal model compounds hard, which is why 62% of TrustRadius reviewers flag the billing as opaque.
- Pricing transparency
- 5.5/10
- Feature depth
- 9/10
- Onboarding speed
- 8.5/10
- Cost at scale
- 5/10
- Agent auto-discovery ships working dashboards within minutes of install, not days
- One query language spans infra, APM, logs, RUM, and synthetics instead of five tools
- G2 rates it 4.4/5 across 884 reviews, the widest install base of any direct rival
- Per-host plus per-signal billing means a 200-host burst month can run 3-4x your baseline
- Custom metrics and high-cardinality tags bill outside the host price and are the top source of surprise invoices
- No self-hosted option exists, so pricing and roadmap changes are entirely Datadog's call
14-day trial, billing starts automatically after
How we priced this
- Tested for
- 6 days
- Plan paid
- Infrastructure Pro ($15/host/mo) and Enterprise ($23/host/mo) -- modeled, not purchased
- Version tested
- Public pricing page + docs, live-captured September 6, 2026
- Test period
- 2026-08-31 → 2026-09-06
This is a pricing analysis, not a 30-day hands-on trial, and we're not going to pretend otherwise. We modeled Datadog's bill at three team sizes (25, 100, and 400 hosts) using the published rate card, screenshotted the pricing page live on September 6, 2026, and cross-referenced every number against Datadog's own docs FAQ. We then pulled aggregate review data from G2 (884 reviews), Capterra (321 reviews), and TrustRadius (354 reviews) to check whether the pricing complaints in public threads matched what verified users report. Finally, we priced the same host counts against New Relic, Grafana Cloud, and Dynatrace using each vendor's own published rate cards, dated the same week. No paid Datadog subscription was used for this analysis; every dollar figure traces back to a public source, cited inline.
Should you pay for Datadog?
YES if you...
- Platform teams under 50 hosts who want a working dashboard the same day, not after a Terraform sprint
- Teams already paying for 3+ separate monitoring tools and want one query language to replace them
- Orgs that can commit annually and negotiate the 15-30% off-list discount vendors quietly offer
NO if you...
- Teams scaling past 200 hosts on a fixed budget -- the per-host math stops being kind
- Anyone who needs self-hosted or on-prem observability for compliance reasons
- Startups with unpredictable host counts -- high-watermark billing charges the whole month at your peak
Datadog's actual rate card
Infrastructure and APM bill on separate lines -- budget for both.
Free
Core collection, capped
- Up to 5 hosts
- 1-day metric retention
- Community support only
Infrastructure Pro
Where most teams start
- 100 custom metrics allotted per host
- 15-month metric retention
- 1,000+ integrations
- Anomaly detection and alerting
Infrastructure Enterprise
Admin controls and governance
- 300 custom metrics allotted per host
- Cross-org visibility, governance console
- SLO tracking built in
APM (add-on)
Distributed tracing, on top of infra
- Code-level distributed tracing
- Correlates traces with logs and RUM
- Not bundled into any infra tier
ROI breakdown: At 40 hosts on Pro + APM, list price is $15 + $31 = $46/host/month, or $1,840/month, before a single log line is indexed. Multi-year and volume discounts kick in past ~500 hosts/month, per Datadog's own pricing page.
Hidden costs & gotchas
- APM bills separately at $31/host/month on top of infrastructure -- it is never bundled
- Log indexing costs $0.10/GB plus roughly $1.27-1.70 per million events ingested
- Custom metrics and high-cardinality tags exceed the per-host allotment and bill on their own line
- High-watermark billing charges the whole month at your peak host count, not a prorated average
What the rate card actually costs
Modeled against Datadog's own published pricing, September 2026.
- Infrastructure Pro
- 15 $/host/month, billed annually Datadog pricing page, captured live
- Infrastructure Enterprise
- 23 $/host/month, billed annually Datadog pricing page
- APM add-on
- 31 $/host/month, billed separately Never bundled with infra tiers
- Log indexing
- 0.10 $/GB indexed Plus $1.27-1.70 per million events ingested
- G2 aggregate rating
- 4.4 /5 across 884 reviews G2 Datadog seller page
- TrustRadius score
- 8.6 /10 across 354 reviews 62% of reviewers flag pricing as opaque
Model a 40-host Infrastructure Pro + APM bill at list price
Check what happens to the bill on a 5-day burst from 40 to 200 hosts
Check what APM alone adds once infra is already paid for
Pros and cons of Datadog's pricing
Pros
- Free tier and 14-day trial remove the risk of committing blind You can run up to 5 hosts free indefinitely, or trial Pro features for 14 days before billing starts -- no card-required guesswork.
- One bill replaces what used to be 3-4 separate vendor contracts Infra, APM, logs, RUM, and synthetics under one invoice is genuinely simpler than reconciling Prometheus plus a log vendor plus a tracing vendor.
- Volume and multi-year discounts are real and negotiable Datadog's own pricing page advertises multi-year/volume discounts past roughly 500 hosts, and public negotiation reports cite 15-30% off list.
- 884 G2 reviews and a 4.4/5 average signal the widest verified install base Among direct competitors, no rival matches that review volume, which makes the aggregate score harder to game than a smaller sample.
Cons
- APM, logs, and custom metrics each bill on a separate line from infra The $15/host headline number never includes tracing or logs. Budget for $40-60/host/month once APM and moderate log volume are in the mix.
- High-watermark billing punishes any burst, even a short one A 5-day scale-up to 200 hosts can price the entire month near the 200-host rate instead of a prorated blend, per Datadog's own billing FAQ.
- 62% of TrustRadius reviewers specifically flag the cost model as opaque That's not a fringe complaint. It's the single most repeated theme across TrustRadius, Capterra's 4.1/5 Value for Money sub-score, and public Reddit threads.
- No self-hosted tier exists, so there's no cost ceiling you control Every other item on this list is a pricing complaint you can budget around. This one means Datadog sets the rules for how the meter runs, permanently.
Final verdict
Datadog's $15/host headline price is real, but it's not the number that shows up on your invoice. Add APM at $31/host, log indexing at $0.10/GB, and the custom metrics that every team ends up needing, and a realistic bill lands at $40-60/host/month for a team running infra plus tracing plus a normal log volume.
Under 50 hosts, that's a fair trade for skipping months of Prometheus-plus-Loki-plus-Tempo integration work. Datadog's own agent gets you a working dashboard the same afternoon, and 884 G2 reviews at 4.4/5 back that up.
Past 200 hosts, the math changes. High-watermark billing means bursts get expensive fast, custom metrics compound outside the host price, and 62% of TrustRadius reviewers already call the model opaque. At that scale, run the numbers against Grafana Cloud's consumption pricing or New Relic's per-GB model before you renew.
Recommended for: platform teams under 50 hosts consolidating from 3+ tools.
Not recommended for: teams past 200 hosts on a fixed budget, or anyone who needs a self-hosted fallback.
- Pricing transparency 5.5/10 Headline price excludes APM, logs, and custom metrics
- Feature depth 9/10 Infra, APM, logs, RUM, synthetics under one query language
- Onboarding speed 8.5/10 Agent auto-discovery, dashboards within minutes
- Cost at scale 5/10 High-watermark billing compounds past ~200 hosts
- Review consensus 8/10 4.4/5 on G2 (884), 4.6/5 on Capterra (321), 8.6/10 on TrustRadius (354)
Common questions about Datadog pricing
How much does Datadog actually cost per month?
Does Datadog have a free plan?
Why is my Datadog bill higher than the advertised price?
What is Datadog's high-watermark billing?
Is Datadog cheaper than New Relic?
Is Datadog worth the price for a small team?
What do real Datadog customers say about pricing?
Update log
- Initial publication: full pricing model at 3 team sizes, 3-platform review aggregation, live pricing screenshots.