Dynatrace pricing

About

Dynatrace is a publicly traded (NYSE: DT) unified observability, security, and business-analytics platform, founded in 2005 by Bernd Greifeneder and spun out as an independent company from Compuware in 2018 following a Thoma Bravo-backed buyout; it IPO’d in August 2019 and is headquartered in Waltham, Massachusetts. The platform combines Full-Stack and Infrastructure Monitoring, Log Management, Digital Experience Monitoring (Real User and Synthetic Monitoring), and Application Security on top of Grail, its purpose-built observability data lakehouse, and Smartscape, its automated topology-mapping engine; Davis AI and the newer generative Davis CoPilot layer causal AI and natural-language querying on top of that data.

Dynatrace competes most directly with Datadog, New Relic, Splunk Observability, Cisco AppDynamics, and Grafana Labs, and sells to organizations of every size — from cloud-native teams metering a handful of hosts to Fortune-500 enterprises running FedRAMP- and HIPAA-regulated estates — while reporting over $1.6 billion in annual recurring revenue. Where most observability incumbents publish a ladder of named plans (Free / Pro / Enterprise), Dynatrace instead sells a single licensing construct — the Dynatrace Platform Subscription (DPS) — that replaces per-product tiers with one annual spend commitment metered against a fully published, per-unit rate card.

Pricing summary : How Dynatrace’s Platform Subscription consumption model works

Dynatrace uses a consumption-based platform commitment model called the Dynatrace Platform Subscription (DPS), with 4 dimensions:

1. One annual commitment, not per-product or per-month: Customers commit to a minimum annual spend at the platform level (not per capability), and that commitment unlocks every generally-available platform capability on day one — there is no separate Enterprise SKU gating advanced features.
2. Rate-card consumption: Every capability draws down from the commitment at a fixed, publicly published per-unit rate — e.g., Full-Stack Monitoring at $0.01/memory-GiB-hour (~$58/mo per 8 GiB host), Infrastructure Monitoring at $0.04/host-hour ($29/mo per host), Log Management ingest at $0.20/GiB, Real User Monitoring at $0.00225/session. More than 30 independently rated capabilities span observability, logs, telemetry, RUM/synthetic, and application security.
3. On-demand overage at the same rate, no penalty: Consumption beyond the annual commit continues to be billed monthly at the identical rate-card price — Dynatrace explicitly states it never charges penalty-style overages; customers can instead raise their commit to unlock a larger volume discount.
4. No per-seat fees: Unlimited user seats are included in every contract regardless of commitment size — cost scales purely with consumption, never headcount.

What makes this different: Dynatrace is one of the few observability vendors that couples a real annual committed-spend contract with a fully transparent public rate card down to the fractional cent, letting buyers price their own bill without a sales call even though the underlying agreement is a commitment, not a self-serve monthly plan.

Pricing by product

Application and Infrastructure Observability

Container Observability

Log Management & Analytics

Telemetry (Metrics, Traces, Events)

Real User & Synthetic Monitoring

Application Security

Sales motions across products: self-serve for the 15-day free trial and for reading list/rate-card prices (published without a sales call); sales-led for sizing and negotiating the annual Dynatrace Platform Subscription commitment itself and for volume discounts on larger commits.

Dynatrace Platform Subscription (DPS): how the annual commitment works

DPS is the only Dynatrace licensing construct, there is no separate self-serve monthly plan. A customer signs an annual order form committing to a minimum spend at the platform level (not per product), which immediately unlocks every capability that is generally available at signing. Usage across all capabilities then draws down against that commitment at the published rate-card price. Per the Dynatrace pricing FAQ: larger commitments earn a lower unit price (as commitment volume increases, unit price decreases), and consumption beyond the committed amount is never penalized, it is billed monthly at the same rate-card price, with the alternative of raising the commit for a better rate. Unlimited seats, SSO, granular access policies, and compliance coverage (ISO 27001, SOC 2 Type II, FedRAMP Moderate, HIPAA) are bundled into every DPS contract regardless of size.

Hidden costs : how the multi-meter rate card actually stacks up

The advertised “$58/mo per 8 GiB host” headline is a per-unit rate, not a bill — a real invoice sums that rate across however many independently metered capabilities a team turns on, and two mechanics inflate it further than the sticker price implies. First, Full-Stack and Infrastructure Monitoring bill by memory-GiB-hour, not by host count, and Dynatrace’s own documentation specifies a 4 GiB minimum per host with RAM rounded up to the next 0.25 GiB increment — a 2 GiB container is billed as if it were a 4 GiB one. Second, most estates layer several add-on meters (Kubernetes Platform Monitoring, Log Management, RUM, Application Security) on top of the base host rate, and each bills independently. Two real-world archetypes built from the published rate card:

Archetype 1 — a 50-node Kubernetes fleet of small (2 GiB) containers

The 4 GiB rounding floor alone adds roughly 25% on top of the naive “50 hosts x $58” estimate before a single add-on capability is switched on — small, container-dense fleets pay a meaningfully higher effective rate per GiB than the headline number suggests.

Archetype 2 — a 300-host enterprise estate with Application Security enabled

Turning on both Application Security modules roughly doubles the per-host observability spend before a single log byte is counted, and choosing Bundled Queries’ higher per-GiB-day retention rate for query predictability becomes the single largest line item once retention windows stretch past a few weeks — the same trade-off Dynatrace’s own Log Analytics section flags when it invites customers to “mix and match your logs pricing model … according to your use-case.” This is exactly the kind of bill unpredictability that a fully published rate card doesn’t eliminate on its own — transparency tells you the unit price, not how many units a real deployment will consume.

Want to estimate your own Dynatrace bill? Use the Dynatrace pricing calculator to model your monthly cost based on host count, memory-GiB, log/trace volume, and which capabilities you turn on.

Pricing evolution : from Host Unit licensing to the Platform Subscription rate card

Cadence

Tracked range: 2022-01–2026-09. 2023 Q1, 2023 Q4, and 2024 Q2 fell on JS-rendered Wayback snapshots that did not preserve pricing content and are marked unknown rather than assumed stable; all other quarters in the tracked range not listed above were confirmed stable.

Notable changes

• 2022 (between March and May) — Dynatrace raised headline module starting prices roughly 5–25% across Full-Stack Monitoring, Infrastructure Monitoring, and Application Security, per Wayback Machine snapshots of dynatrace.com/pricing.
• 2023-04-26 — Dynatrace launched the Dynatrace Platform Subscription (DPS), replacing per-module Host Unit/DDU/DEM Unit pricing with a single annual platform-level commitment metered against a published rate card; source: Dynatrace’s own announcement blog.
• 2023-06-27 — A follow-up Dynatrace blog post confirmed Infrastructure Monitoring under DPS removed host-memory limits and moved to 15-minute-interval billing.
• 2024 (by September) — Kubernetes Platform Monitoring became a standalone rate-card line item at $0.002/hour per pod.
• 2024-10 — Davis CoPilot, Dynatrace’s generative-AI assistant, reached general availability and was made available to Platform Subscription customers without an added rate-card line item.
• Between late 2024 and the September 2026 baseline capture (exact date unknown) — Application Security was unbundled into two separately metered capabilities (Runtime Vulnerability Analytics and Runtime Application Protection), and a new Foundation & Discovery entry tier was introduced; together these roughly double the cost of Application Security coverage that was a single SKU at DPS launch.

What’s unique : distinctive mechanics behind Dynatrace’s rate card

1. A real annual commitment married to a fully public per-unit rate card. Most enterprise-observability vendors force a choice between self-serve transparency (a published price list, usually capped at SMB-scale plans) and negotiated enterprise pricing (a real commitment, but sales-only rates). Dynatrace publishes 30+ rate-card line items to the fractional cent while still requiring a signed annual spend commitment — buyers can model their own contract before ever talking to sales, then negotiate volume discounts from a known baseline instead of a blind quote.

2. Memory-based host billing, not host-count billing. Full-Stack and Infrastructure Monitoring aren’t priced “per server” the way most APM competitors price hosts — they’re priced per GiB of RAM the monitored host actually carries, at hourly granularity, with a 4 GiB minimum floor. A 64 GiB database host costs roughly 8x what an 8 GiB microservice host does under the identical rate, which tracks real telemetry volume more closely than flat per-host pricing but also means capacity sizing, not headcount, drives the bill.

3. Zero seat fees at any commitment size. Unlike Datadog’s per-host-plus-per-user add-ons or New Relic’s per-user “Core”/“Full Platform” seat tiers layered on top of ingest, Dynatrace charges nothing for user accounts regardless of contract size. That removes the common friction of rationing dashboard access to control license cost — the entire economic surface is consumption, not headcount.

4. Deterministic Davis AI is a platform primitive, not an upsell. Davis, Dynatrace’s causal (non-generative) AI engine for automated root-cause analysis, ships bundled into every monitoring capability rather than gated behind a premium tier or billed as its own SKU. The newer generative layer, Davis CoPilot, reached general availability in October 2024 and is likewise made available to any Dynatrace Platform Subscription customer rather than sold as an added-cost line item on the public rate card — a structural difference from vendors that price AI-assisted analysis as a separate add-on.

5. No penalty-style overages, only continued consumption at the same rate. Where many usage-based SaaS platforms multiply the unit price once a customer exceeds committed capacity, Dynatrace’s stated policy is to keep billing overage at the identical rate-card price indefinitely — the only consequence of under-committing is missing out on the volume discount a larger commitment would have unlocked.

Strengths & weaknesses

Billing UX : Named controls on Dynatrace pricing surfaces

• View full rate card / View FAQs links - repeated under nearly every product section on the main pricing page, letting a buyer jump straight from a summary price to the full per-unit rate-card table or the FAQ tab for that capability.
• Full Rate Card page (dynatrace.com/pricing/rate-card) - a single scrollable Category > Capability > List Price (USD) table covering 30+ SKUs, from Full-Stack Monitoring down to AppEngine Functions and legacy Platform Extensions Classic units, plus a Contact us for available discounts button and a View currency exchange rates link at the foot of the table.
• Tabbed FAQ (How the model works / How usage is measured / Additional details) - splits DPS mechanics into three separate tabs on both the main pricing page and the dedicated FAQ page rather than one long scroll; only the How the model works tab content was captured this pass (the other two tabs were not actuated).
• Local Currency Pricing table (dynatrace.com/pricing/currency-conversion) - a dedicated page listing 18 supported non-USD currencies (Euro, British Pound, Japanese Yen, Indian Rupee, Brazilian Real, and more) with the exact fixed multiplier Dynatrace applies to its USD list prices.
• Built-in cost-monitor and cost-allocation controls - referenced directly on the Platform Subscription page as automated forecasting, alerts on sudden usage changes, and cost allocation by team/product, each linking out to dedicated docs rather than being described inline.
• Asterisked per-unit rate footnotes - every flat monthly headline price on the main pricing page (e.g. Foundation & Discovery at $7/mo) carries an asterisk translating it into the true rate-card unit (e.g. *Billed at $0.01 per hour per host), so the hourly/volume mechanic is always one line below the headline number.

Strategic wins : why specific DPS decisions worked

1. Publishing a full per-unit rate card instead of hiding behind “Contact us”

Most enterprise observability vendors — Splunk, AppDynamics, and historically Dynatrace itself — keep the actual unit economics behind a sales quote. Publishing 30+ rate-card line items to the fractional cent gives buyers a self-serve way to model total cost before a single sales conversation, a pattern more common in usage-based pricing among developer-tools vendors than among legacy APM incumbents.

2. Replacing “high watermark” monthly licensing with true hourly consumption

Before DPS, Dynatrace billed hosts monthly regardless of how long they actually ran; the 2023 shift to hourly billing (tracked in 15-minute intervals) means ephemeral, autoscaled, or short-lived infrastructure no longer inflates the bill for capacity that only existed for a fraction of the month — a meaningful usage-metering improvement for cloud-native customers, and one worth reading alongside how the migration itself played out for SaaS vendors moving to usage-based pricing.

3. Eliminating per-seat fees entirely

Where competitors often ration advanced-tier seats to control license cost, Dynatrace’s zero-seat-fee model removes the incentive to restrict dashboard access, encouraging the wider cross-team adoption that observability tooling needs to actually catch incidents before customers do.

4. Bundling Davis causal AI and the Grail data lakehouse into the base commitment rather than upselling them

Competing platforms increasingly gate AI-assisted analysis or unified log/metric/trace storage behind premium add-ons; Dynatrace ships both as platform primitives available on day one of any DPS contract, reinforcing the “every capability, unlocked day one” pitch at the center of its pricing summary.

Areas to improve : specific gaps with proposed fixes

1. Reduce the number of independently priced capabilities a buyer has to reason about

30+ separate meters (host-hour, memory-GiB-hour, session, pod-hour, workflow-hour, and more) is a lot of surface area to model, even with every rate published. Consolidating adjacent meters — for example, the repeating three-part ingest/retain/query split across Logs, Traces, Metrics, and Events — into fewer composite units would preserve transparency while cutting cognitive load, closer to the fewer-dimensions ideal in the KISS pricing framework.

2. Surface rounding and minimum-billing rules on the pricing page itself, not just in docs

The 4 GiB memory-rounding floor and 15-minute minimum billing interval materially change what a container-dense fleet pays relative to the advertised per-host rate, but a buyer has to dig into Dynatrace’s documentation to find them. Surfacing them inline — the way the asterisked “*Billed at $0.01/hour” footnotes already do for the base rate — would close a real expectation gap that shows up repeatedly in third-party reviews.

3. Offer a true self-serve option below the annual-commitment floor

DPS still requires a signed annual order form even for a single monitored host. A genuine trial-to-paid self-serve path — as Datadog and New Relic both offer at small scale — would let solo developers and very small teams adopt Dynatrace without a sales cycle, capturing bottoms-up adoption the current model structurally excludes.

4. Give the built-in cost-allocation tooling more visibility in the public pricing narrative

The cost monitors and team/product cost-allocation features are a genuine differentiator for FinOps buyers, but they’re referenced only briefly on the Platform Subscription page rather than demonstrated with real screenshots or a dedicated walkthrough. Clearer documentation here would let Dynatrace compete more directly on the AI cost governance and chargeback narratives that FinOps teams increasingly expect from any consumption-priced platform.

Key takeaways

1. Publishing a full rate card does not require abandoning committed-spend contracts. Dynatrace proves transparency and commitment aren’t mutually exclusive — a vendor can require a real annual minimum while still disclosing every unit price to the fractional cent, letting buyers self-serve the modeling even when the deal itself is sales-led.
2. Removing per-seat fees changes the adoption calculus more than most tiering decisions do. By making user seats free at any size, Dynatrace eliminates the “who gets a license” rationing conversation that slows internal rollout for competitors with per-user pricing — a lesson for any platform trying to drive wall-to-wall internal adoption rather than optimize per-seat revenue.
3. “No penalty overages” is a powerful trust signal worth stating explicitly. Dynatrace’s FAQ answer that it “never charges penalty-style overages” directly counters the most common usage-based-pricing fear — bill shock. Other usage-based vendors should consider stating their overage policy just as explicitly rather than leaving it to be discovered at invoice time.
4. Granular metering creates its own transparency paradox. Publishing 30+ independent rates is more honest than a black-box quote, but it can still leave buyers unable to predict their bill — a reminder that transparency and simplicity are different design goals, and satisfying one doesn’t guarantee the other.
5. Rounding and minimum-billing rules belong next to the headline price, not just in the docs. The 4 GiB memory floor and 15-minute billing interval are exactly the kind of mechanic that erodes trust when a customer discovers it on an invoice instead of the pricing page — any usage-metered vendor should surface its rounding/minimum rules as prominently as its base rate.

UBP implications

1. Consumption-based platform commitments are becoming the enterprise-observability default, not the exception. Dynatrace’s shift from host-unit licensing to the Dynatrace Platform Subscription mirrors a broader pricing-model trend of infrastructure vendors moving from static SKUs to a single spend commitment metered against granular, published consumption — a middle path between pure self-serve usage pricing and traditional enterprise license negotiation.
2. Rate-card transparency at enterprise commitment sizes is a competitive differentiator, not just an SMB tactic. Dynatrace demonstrates that full per-unit price disclosure is compatible with six- and seven-figure annual contracts, challenging the assumption that enterprise deals must stay opaque to preserve negotiating leverage.
3. Multi-meter consumption pricing needs an explicit “no overage penalty” commitment to earn trust. As more usage-based vendors adopt commitment-plus-rate-card models, Dynatrace’s explicit promise that overage bills at the same rate — not a punitive multiplier — sets a bar other consumption-pricing vendors will increasingly be measured against.

Sources

• Dynatrace pricing page (accessed 2026-09-01)
• Dynatrace pricing rate card (accessed 2026-09-01)
• Dynatrace Platform Subscription (DPS) (accessed 2026-09-01)
• Dynatrace pricing FAQ (accessed 2026-09-01)
• Dynatrace currency conversion (accessed 2026-09-01)
• How Dynatrace pricing works (docs) (accessed 2026-09-01)
• Full-Stack Monitoring consumption calculation (docs) (accessed 2026-09-01)
• Introducing the Dynatrace Platform Subscription (blog) (accessed 2026-09-01)
• The Dynatrace Platform Subscription model enables broad Infrastructure Monitoring (blog) (accessed 2026-09-01)
• Announcing General Availability of Davis CoPilot (blog) (accessed 2026-09-01)

Bottom line

Dynatrace’s Platform Subscription is a rare enterprise-software artifact: a fully published, fractional-cent rate card wrapped around a real annual commitment, with zero seat fees and an explicit no-overage-penalty promise — trading some bill predictability for a level of pricing transparency most observability incumbents still refuse to offer.

Want to compare Dynatrace against other observability pricing? Browse the pricing blueprint.

Pricing timeline : Major events on a vertical axis

Each milestone below corresponds to a public pricing change, product launch, or material adjustment. Major events use a filled marker; minor adjustments use a faded one.

Baseline capture: Dynatrace Platform Subscription (DPS) consumption rate card

Sep 2026

Full-Stack Monitoring $58/mo per 8 GiB host ($0.01/memory-GiB-hour), Infrastructure Monitoring $29/mo per host ($0.04/host-hour), and Foundation & Discovery $7/mo per host ($0.01/host-hour) are now discrete tiers, and Application Security is split into two separately billed meters — Runtime Vulnerability Analytics and Runtime Application Protection, $13/mo per 8 GiB host each — versus the single $0.018/hour bundled Application Security SKU sold at DPS launch in 2023. The rate card now spans 30+ independently priced capabilities with no per-seat fees and no penalty overages.

Infrastructure Monitoring: unlimited host memory, hourly billing formalized

Jun 2023

Dynatrace published a follow-up blog post confirming Infrastructure Monitoring under DPS removes prior host-memory limits and bills in 15-minute intervals rather than by monthly host count, enabling broader deployment across large, autoscaled IT environments. Source: dynatrace.com/news/blog, June 27, 2023.

Dynatrace Platform Subscription (DPS) launch — model transition from host units to hourly rate-card consumption

Apr 2023

Dynatrace announced the Dynatrace Platform Subscription, replacing the Host Unit / DDU / DEM Unit licensing model with a single annual platform-level spend commitment metered against a published hourly rate card (Full-Stack Monitoring $0.08/hour per 8 GiB host, Infrastructure Monitoring $0.04/hour for any host size, Application Security $0.018/hour per 8 GiB host as one bundled SKU). Dynatrace states the model eliminates 'high watermark' monthly licensing and per-user fees. Source: dynatrace.com/news/blog, April 26, 2023; confirmed live on the pricing page in the May 2023 Wayback snapshot.

Mid-2022 module price increase

May 2022

Between the March 2022 and May 2022 Wayback snapshots, Dynatrace raised its headline starting prices: Full-Stack Monitoring $69 to $74/mo per 8 GB host (+7%), Infrastructure Monitoring $21 to $22/mo (+5%), and Application Security $12 to $15/mo (+25%). Digital Experience Monitoring ($11/mo) and Open Ingestion ($25/mo per 100k DDUs) held steady. Rates remained stable at the new level through at least the October 2022 snapshot.

Pre-DPS Host Unit / DDU / DEM Unit pricing (Wayback baseline)

Jan 2022

Under the legacy licensing model, Dynatrace sold module-based starting prices: Full-Stack Monitoring $69/mo per 8 GB host, Infrastructure Monitoring $21/mo per 8 GB host, Digital Experience Monitoring $11/mo per 10k annual DEM Units, Application Security $12/mo per 8 GB host (add-on), and Open Ingestion $25/mo per 100k annual Davis Data Units (DDUs). Source: web.archive.org snapshot of dynatrace.com/pricing, January 2022.