SAP Observability Tools Compared

by | Sep 16, 2026

SAP observability comparison

Updated November 2026

Comparisons of SAP observability tools often evaluate which platforms can see inside SAP. Today, that’s nearly all of them. Dynatrace, Datadog, New Relic and Splunk can all get SAP telemetry. None of them are likely the best choice for an SAP-centric application, and we will document why.

Three questions to ask first

1

Where does the SAP depth come from?

First-party engineering, or a separately licensed add-on from a partner?

2

What was the platform built to answer?

Tools carry their heritage into every report they email and every dashboard they draw.

3

Can it act, or only watch?

Knowing an update process is stuck is worth far less than an automated resolution.

That last question is where most of these platforms stop, and it is the difference between observability and operations.

Heritage matters

When Basis teams evaluate observability with an SAP application in mind, the primary focus should be “what do we need to keep SAP running?” Generic IT monitoring solutions, robust as they are, were often not designed around this specific task. This compromises SAP operations, just as attempting to use Solution Manager for general enterprise service management would leave a lot of uncovered ground outside SAP.

Dynatrace · New Relic · AppDynamics

Bytecode instrumentation

Dynatrace, New Relic and AppDynamics grew out of the bytecode instrumentation of Java and .NET applications. Their native instinct is the distributed trace: follow a request through the stack and find the slow span.

Splunk

Machine data

Splunk grew out of machine data. Its instinct is to index everything and ask questions afterwards, which is why it became a SIEM before it became an observability platform.

Datadog

Cloud infrastructure

Datadog grew out of cloud infrastructure monitoring, and its instinct is breadth: a thousand integrations, one pane of glass, agents everywhere.

Avantra · IT-Conductor

SAP Basis work

IT-Conductor and Avantra came from SAP Basis work. Their instinct is the daily check: is the update queue clean? Did the batch chain finish? Is the enqueue server healthy?

None of these approaches are inherently wrong. They answer different questions, which is why the comparison isn’t really a ranking. You don’t ask whether a ratchet drive is better than a screwdriver. You ask “what tool do I need for this job?”

For most enterprises that job is this: SAP runs the processes that determine whether the quarter closes, the orders ship, and people get paid. Everything else in the estate is being monitored so that those things keep happening.

At a glance

Swipe the table sideways to see every column.

Tool Heritage SAP depth Automation Pricing basis Best for
Avantra SAP Basis operations First-party, purpose-built Yes
Runbook and workflow automation across the landscape
Avantra Licensed Objects SAP-first operations teams who want to fix, not just find
IT-Conductor SAP Basis managed services First-party, agentless via RFC/CCMS Yes
Automated recoveries and workflow orchestration
Per system per month, tiered Smaller SAP estates and orchestration across SAP and non-SAP
Dynatrace Bytecode instrumentation PowerConnect (Rhondos), a separately licensed SAP-certified add-on, plus first-party ABAP/HANA/JVM extensions No
Analysis and alerting, remediation elsewhere
Per host plus Davis Data Units, PowerConnect licensed separately Enterprises standardizing full-stack observability with SAP included
Datadog Cloud infrastructure monitoring Redpeaks (formerly Agentil) and RapDev marketplace apps, plus first-party SAP HANA integration No Per host, plus marketplace licenses Cloud-native estates where SAP is one workload among many
New Relic Bytecode instrumentation First-party, SAP-certified ABAP add-on on a central collector No Consumption, plus an add-on charge per SAP SID Teams already standardized on New Relic across non-SAP apps
Splunk Log and machine data analytics Splunk AppDynamics Monitoring for SAP Solutions, plus log ingestion No Ingest volume and workload pricing SIEM-led estates where security and operations share a data platform

For Dynatrace, Datadog, New Relic and Splunk, the deep SAP capability is supplied by a product licensed separately, from a different vendor, with its own prerequisites and its own support queue. Only the two platforms with SAP Basis heritage close the loop from detection to action.

Replacing CCMS and NetWeaver Administrator

For many Basis teams, SAP monitoring still means CCMS for ABAP systems, with the RZ20 alert monitor and RZ21 thresholds, and NetWeaver Administrator for SAP Java systems.

Avantra works without CCMS. Its built-in checks cover the areas CCMS monitors in ABAP systems, from work processes, updates and the lock table to short dumps, background jobs, spool, RFC queues and IDocs, as well as the operating system and database. SAP Java systems are monitored natively too. If you have years of CCMS monitors you don’t want to lose, Avantra can bring them in as checks and import their alerts, so you can retire CCMS at your own pace.

What replaces what

Legacy function Where it lives today Avantra
ABAP alert monitoring CCMS alert monitor (RZ20) Built-in checks for work processes (WorkProcesses), updates (UPDATEQUEUE), the lock table (LOCKTABLE), short dumps (SHORTDUMPS), background jobs (JOBS), spool (SPOOL), RFC queues and IDocs (TRFC, QRFC, IDOCS) and buffers (SAPBufferSummary)
Thresholds CCMS properties (RZ21), maintained per system Parameter Sets that apply thresholds to whole groups of systems, with optional per-system overrides
Operating system data CCMS agents The Avantra Agent, with built-in checks for CPU, memory, file systems and paging space
Existing custom monitors CCMS monitoring tree elements and monitor sets CCMS and CCMS_MONSET custom checks, which bring them into Avantra as real-time checks
CCMS alerts CCMS alert browser Imported into Avantra, optionally confirmed in SAP, and sent by email, to ServiceNow or to a REST endpoint. Notifications can also start automation workflows (Automation, Cloud and Enterprise Editions)
Java availability NetWeaver Administrator J2EECONNECT (over the same P4 port NetWeaver Administrator uses), J2EEConnectHttp and JavaServerNodes
Java performance, services and jobs NetWeaver Administrator Thread pool, JVM memory, garbage collection and response time checks, J2EE_COMPONENT_STATUS for applications and services, JavaJobStat for Java Scheduler jobs, and J2EE_JMX_MON for the AS Java monitor tree
Java certificates and configuration NetWeaver Administrator JavaSSLCertificatesValidity, which warns as certificates approach expiry, and change detection for Java properties, patch level and software components

This covers NetWeaver Administrator’s monitoring role. For administration tasks such as deployment, configuration and user management, keep using NetWeaver Administrator.

For a wider plan to move off Solution Manager-era tools, see SAP Solution Manager transition with Avantra.

Heritage: SAP Basis managed services

IT-Conductor

Known for: an agentless, SaaS-delivered SAP monitoring and orchestration platform, with automation reaching into transports, system refresh and IT workflow.

What it does

  • Agentless collection from ABAP and Java stacks via RFC, CCMS and custom function modules across ECC, S/4HANA, BW/4HANA and Solution Manager
  • Availability monitoring through synthetic checks and an external heartbeat
  • Static, dynamic and standard-deviation thresholds with role- and user-based overrides
  • Graphical workflow orchestration spanning SAP and non-SAP systems, with hooks into Ansible and Terraform
  • Automated recoveries, transport automation and system refresh tooling

The trade-off in pricing: IT-Conductor publishes its rates, which is refreshing. The entry paid tier covers SAP NetWeaver, BOBJ and cloud apps. Native database and OS monitoring starts at the next tier up. The free tier is capped at two systems, application layer only, with no custom monitors or workflows beyond the built-in templates.

Nothing to install is a real advantage during procurement and deployment. It can become a real limitation when visibility into the operating system and database is limited and the issue lies below the application layer.

Agentless collection also depends on the managed system being reachable and responsive enough to answer, a condition which can fail during critical incidents.

Pros

  • Nothing installed on managed SAP systems, which shortens security review
  • Genuinely spans SAP, cloud and non-SAP infrastructure
  • Published pricing, with a free tier for small estates
  • Real automation, not just alerting

Cons

  • Depth at the OS and database layers is a paid upgrade, not a baseline
  • Collection depends on the managed system responding to remote calls
  • Broad coverage is assembled from several separately named modules rather than one platform
  • End-user monitoring is priced separately again
PricingPer system per month, tiered by monitoring scope, discounted by contract term.
Heritage: bytecode instrumentation

Dynatrace

Known for: AI-driven full-stack observability, with SAP brought in through PowerConnect.

What it does

  • First-party observability for SAP Commerce Cloud, SAP JVM, SAP ABAP via extensions, and remote monitoring for SAP HANA
  • PowerConnect, an SAP-certified add-on authored by Rhondos and installed into the SAP system, collecting from monitoring transactions, APIs, function modules, IDocs, reports, change documents and CCMS, and covering Cloud ERP Private, BTP and SAP SaaS applications
  • SAP telemetry landed in the Grail data lakehouse, with over 200 prebuilt SAP dashboards
  • Davis AI correlating SAP response times with dependencies across web front ends, databases and cloud services
  • Automated topology mapping and root cause detection across tiers

What to weigh: the SAP depth is genuine, and it belongs to PowerConnect, a second vendor licensed separately from Dynatrace. It is installed into the SAP stack under Basis change control and upgraded on its own cadence. Two contracts, two support paths, two roadmaps.

Dynatrace also observes; it does not operate. There is no runbook executing a system refresh or clearing a stuck update queue at the end of the analysis.

Pros

  • Best-in-class correlation and automated root cause analysis across the full stack
  • Strong hybrid and multi-cloud coverage well beyond SAP
  • Mature ecosystem, with SAP dashboards that practitioners recognize

Cons

  • SAP depth depends on a separately licensed partner product
  • Per-host plus Davis Data Unit pricing climbs quickly at enterprise scale, before PowerConnect is added
  • Analysis stops at the alert; remediation lives somewhere else
PricingPer host plus consumption-based Davis Data Units. PowerConnect licensed separately.
Heritage: cloud infrastructure monitoring

Datadog

Known for: breadth. Over a thousand integrations and a single view across a cloud estate.

What it does

  • A first-party SAP HANA integration covering memory, disk, and data and log volume growth
  • An SAP S/4HANA and NetWeaver integration powered by Redpeaks, formerly Agentil Pro.Monitor, delivered through the Datadog Marketplace: agentless remote collection with preconfigured templates for short dumps, SAP jobs, transaction response times and work processes, supporting on-premise, private cloud, hybrid and RISE
  • A RapDev integration pulling SAP Cloud ALM health, job, exception and integration monitoring into Datadog over OpenTelemetry
  • Watchdog anomaly detection across whatever is ingested

What to weigh: Datadog’s SAP coverage is better than its reputation among Basis teams suggests, but almost none of it is native Datadog. Redpeaks is a marketplace product with its own collector to deploy per data center, its own license, its own prerequisites and its own support address.

For an organization already standardized on Datadog, that may be exactly the right approach. It is a weaker fit for customers with large existing on-premises and hybrid environments managed with a mix of tools.

Pros

  • Fast time to value for teams already running Datadog everywhere else
  • Genuinely unified view across SAP and the surrounding cloud estate
  • One of the broadest integration catalogues in the market

Cons

  • SAP depth is supplied by third-party marketplace products, licensed and supported separately
  • Per-host pricing scales with every node across a large landscape
  • No automation reaching back into the SAP system
PricingPer host, plus separate marketplace licenses.
Heritage: bytecode instrumentation

New Relic

Known for: full-stack observability with a first-party, SAP-certified SAP integration.

What it does

  • New Relic Monitoring for SAP Solutions: an SAP-certified ABAP add-on installed on a single central monitoring system, pulling telemetry from the rest of the landscape over RFC and forwarding it over HTTPS
  • SAP KPIs including dialog response time, transaction volume, job execution statistics and RFC performance
  • BTP monitoring through SAP Cloud ALM OpenTelemetry APIs
  • NRQL for custom SAP dashboards and alerting
  • Correlation between SAP telemetry and the rest of the application estate, including Fiori and UI5 front ends

What to weigh: New Relic owns its SAP integration rather than reselling one, which puts it ahead of Dynatrace and Datadog on vendor simplicity. The architecture is the same central-collector, agentless pattern used by IT-Conductor, with the same trade-off: the depth available is the depth a remote call can retrieve, and the collector becomes a single point of dependency for the landscape.

Like the others in this group, it reports and correlates, with no automated action capability.

Pros

  • First-party SAP integration, SAP-certified, no third-party license in the path
  • Agentless, which shortens deployment into production systems
  • Strong correlation between SAP and the surrounding application portfolio
  • RISE with SAP support

Cons

  • Consumption pricing plus a per-SID add-on charge makes forecasting harder
  • Central collectors concentrate both load and risk
  • No remediation, and no SAP operations content beyond telemetry
PricingConsumption-based, with an add-on charge per SAP System ID.
Heritage: logs and machine data

Splunk

Known for: the SIEM and log platform that acquired serious APM capability when Cisco folded AppDynamics into Splunk in 2025.

What it does

  • SAP log ingestion and correlation through forwarders and HTTP Event Collector
  • Splunk AppDynamics Monitoring for SAP Solutions, a named product in the Splunk Observability portfolio covering application and business performance monitoring for SAP
  • Prebuilt SAP security dashboards covering logins, authorizations, systems and connections
  • Business transaction tracing and code-level diagnostics inherited from AppDynamics, though not comparable in detail to bytecode instrumentation for J2EE
  • SIEM and compliance reporting on the same data platform as operational monitoring

What to weigh: Splunk’s SAP story is real and underestimated, particularly the security angle, where sharing a platform between SecOps and SAP operations has obvious appeal. The caution is commercial rather than technical: the portfolio is mid-consolidation across Splunk Platform, Splunk AppDynamics, Observability Cloud and ITSI. Ingest-based pricing and SAP log volume are not natural friends for buyers.

Pros

  • Security and operations on one data platform
  • Genuine APM capability with a dedicated SAP product
  • Scales well where logs are already centralized in Splunk

Cons

  • Ingest-volume economics deteriorate as SAP log volume grows
  • Portfolio consolidation still in progress across four product lines
  • Operations remains a reporting function, not an execution one
PricingIngest volume and workload-based.

How we evaluated these platforms

We dropped the criterion most comparisons lead with, because it does not survive contact with an SAP system. ABAP usage data from UPL and its successor SCMON is often presented as deep application visibility. It isn’t. It records which ABAP units executed, and in SCMON’s case, what called them. It carries no timing and no call tree. It is activated deliberately with a record limit and a scheduled deactivation, purged unless aggregated through SUSG, and SAP advises against running it on systems already at maximum load. Its real consumers are custom code housekeeping, clean core assessment and test scoping. It is not an operational monitoring capability, and no platform in this comparison should be credited or faulted for it.

What we evaluated instead:

  • Built-in operational depthWork processes by type, update and enqueue health; short dumps, batch chains, spool, RFC and IDoc available out of the box, or after a template-building project
  • Detection methodStatic thresholds, or baselines learned per system
  • Detection to actionWhether the platform can execute remediation inside the landscape or only raise an alert for a human
  • Coverage anywhere SAP runsOn-premise, hyperscaler, RISE with SAP and Cloud ERP Private, BTP, S/4HANA Cloud
  • SAP operations content beyond metricsHotNews and security note analysis, configuration drift, landscape inventory, service-level reporting
  • Acquisition modelWhether SAP depth is first-party or a separately licensed add-on, and what that means for support and upgrades
  • Non-SAP estate coverageWhere the generalists legitimately win

How to choose

The advice you will usually read is that if SAP is one application among many, choose a generic platform, and if you are mostly SAP, choose an SAP-native one. That advice makes a wrongheaded assumption.

App count doesn’t determine risk. Business criticality does. If the general ledger runs on SAP, it does not matter that four hundred other applications run elsewhere. SAP needs first-class operational attention, and infrastructure-layer metrics from a platform treating SAP as one integration among a thousand will leave a blind spot exactly where the business is most exposed.

The realistic answer for most enterprises is not one tool; it is a deliberate decision about which platform owns and ensures SAP availability.

Our recommendation for evaluation:

  1. Establish what runs on SAP and what happens if it fails.Not how many systems: what processes.
  2. Decide who owns SAP operations.If SAP carries critical processes, that owner needs SAP-native depth and, ideally, the ability to act on what it sees.
  3. Determine where SAP runs, today and tomorrow.If you are planning a migration, how much of your estate will be in transition and for how long? Where is the destination? For complex organizations, the answer is often “years” and “across everything,” so evaluate with this in mind.
  4. Decide who owns the rest of the estate,and how the two correlate: shared dashboards, a common alerting layer, or an ITSM integration.
  5. Count the contracts.If the SAP depth in your chosen platform comes from a partner add-on, you have two vendors, not one.

If SAP genuinely carries nothing business-critical in your organization, a generic platform with a connector will do. That is a much smaller set of organizations than the phrase “mixed estate” implies.

Frequently asked questions

How do you replace SAP CCMS monitoring?

Avantra works without CCMS. Its built-in checks cover work processes, updates, the lock table, short dumps, background jobs, spool, RFC queues, IDocs, buffers and the operating system. Thresholds are set in Parameter Sets that apply to whole groups of systems, rather than in RZ21 on each system. To avoid a hard cutover, Avantra can import existing CCMS monitoring tree elements, whole monitor sets and CCMS alerts, so you can retire CCMS at your own pace.

What replaces SAP NetWeaver Administrator for monitoring SAP Java systems?

Avantra monitors SAP Java systems natively. It checks instance and server node availability, thread pools and long-running threads, JVM memory and garbage collection, response times, the status of applications and services, Java Scheduler jobs, SSL certificates approaching expiry, and objects in the AS Java monitor tree. It also detects changes to Java system and VM properties, patch levels and software components. For administration tasks such as deployment, configuration and user management, keep using NetWeaver Administrator.

What are the best observability tools for SAP environments?

It depends on whether SAP runs mission-critical business processes, not on how many other applications sit in the estate. SAP-first platforms like Avantra and IT-Conductor are built around SAP’s operational model (work processes, batch chains, update and enqueue health) and close the loop from detection to automated remediation. Dynatrace, Datadog, New Relic and Splunk offer broader coverage across a mixed estate with SAP depth layered in, in three of those four cases through a separately licensed partner product. If a critical process like the general ledger or order-to-cash runs on SAP, native depth matters regardless of app count.

What is the difference between SAP observability and SAP operations?

Observability tells you what is happening and, with good correlation, why. Operations includes doing something about it: clearing the queue, restarting the job, refreshing the system, applying the note. Every platform in this comparison does the first. Two of them do the second.

How do you monitor SAP across RISE, on-premise and BTP together?

Look for a platform that treats all three as the same operating problem rather than three integrations. The practical test is whether one alerting model, one service-level report and one automation library cover systems wherever they run, or whether RISE, on-premise and BTP each end up with their own console and their own conventions.

Is UPL or SCMON data useful for monitoring?

Not for operational monitoring. UPL is obsolete and its successor SCMON records which ABAP units executed and what called them, without timing or call-path detail. Collection is switched on deliberately, constrained by a record limit policy, and automatically purged unless aggregated through SUSG.