The big 3 observability tools: Datadog vs New Relic vs Splunk

When we consider adding observability to our stack, we may default to popular options like Datadog. While they have a ton of features covering most use cases, they also come with premium pricing attached to it. This blog covers a tool-wise comparison of these platforms, comments from existing users, and our recommendations for your observability stack.

Not only are observability tools costly, costing about 30% of a company’s outside vendor spending, but they also lack visibility and transparency in their pricing, making it nearly impossible to calculate how their pricing has been calculated. The various services have different pricing models, each with limits (number of hosts/devices/requests/ingest) and multiple tiers with monthly and annual billing options.

Data from Statista shows that the Observability tools market, worth $12.9 Billion in 2020, is set to reach $19.3 Billion by 2024. These tools have applications in various industries, from media, communications, financial services, technology, health, the public sector, manufacturing, etc., in all regions of the world. Organizations are also increasing the number of observability tools used while consolidating the number of vendors. (Splunk, State of Observability 2022)

Here is what users have to say

Before we dive into feature-specific comparison, let’s look at what the users of these tools have to say. These are some common likes, dislikes, and good to know we found from over half a dozen review sites, including G2, Gartner, and software advice.

New Relic

Liked:

• Straightforward setup
• Share dashboards externally
• Transparent pricing with great ROI
• AI-based anomaly detection and alerting
• Built on OpenTelemetry standards
• Go backward in time and analyze historical bottlenecks and consumption trends

Disliked:

• Complex query builder, limited regex, and other limitations of NRQL
• The learning curve to make use of all customization options can be overwhelming
• Incorrect query results, bugs, sluggish experience

Keep in mind:

• It comes with a generous free tier; use it yourself to see if it works
• Their network monitoring doesn’t yet support all infra
• Use their data ingest cost estimator

Splunk

Liked:

• Works well for high-volume data ingestion
• Indexing and ML applied to the data make it valuable
• Built-in reports and dashboards that can be customized
• Built on OpenTelemetry standards
• They offer workload, ingest, and entity pricing options

Disliked:

• Monthly pricing is expensive and opaque
• Lack of visualization options, room for improvement with their interface

Keep in mind:

• There is a 10k session/month per host limit for RUM
• If you don’t want the bundle offerings in Standard and Pro plans, you can also choose individual offerings as per your requirements
• Consider Splunk’s flexible pricing options to find the one that best suits you
• If you’re also considering using Splunk for IT Ops and Security, check their cloud platform

Datadog

Liked:

• Stability and the constant addition of new features
• Out-of-the-box integrations
• AI and ML Capabilities
• All tools in one place, with a ton of customization options

Disliked:

• Pricing is opaque, inflexible billing, and unexpected charges
• Hard to find the right metric and look at historical data
• Deploying the agents is still a very manual task

Keep in mind:

• Leverage estimated usage and set alerts to avoid overspending
• Monthly billing is 20-50% higher than annual billing
• Their security offerings are relatively new and not up to speed with the rest of their monitoring
• Their free trial is limited compared to their competitors

Moving into the comparison, we compare each tool by its features. Starting with APM & log management and ending with alerting & network monitoring.

Note: Here we only compare non-enterprise tiers. Free tiers (where available), Splunk’s Standard tier, Datadog’s Pro tier, and New Relic’s Pro tier. If you’re interested in enterprise pricing, check out their pricing pages - Datadog, Splunk, New Relic.

APM - Application Performance Monitoring

Log management

Infra monitoring

Serverless monitoring

Real user management

Synthetic monitoring

Alerts and incident management

Splunk On-Call was previously VictorOps, which was acquired by Splunk in 2018.

Network monitoring

• New Relic’s solution is based on the ktranslate docker container. This single container image is hosted in your environment to collect and process your data to be exported to the Event, Metric, and Log APIs and displayed in New Relic.

---

As we can see from all of the above charts, Splunk and New Relic have nearly caught up to the offerings of Datadog on most fronts. Through their recent push towards observability, recent acquisitions, contributions to open-source projects, and partnerships. They also offer aggressive pricing, which undercuts Datadog’s and other expensive competition on most fronts.

Choosing your observability stack

As the business impact of outages rises day by day, more and more businesses are likely to spend on Observability tools. While the above comparisons give you a basic overview of which tools to use and how much each might cost, here are a few considerations to keep in mind while choosing the observability tool.

1. Analyze which parts of your stack needs monitoring the most, then analyze and try out tools specific to that need. There are open-source and free tiers tools available for most of these features.
   1. Infrastructure
   2. Networks
   3. Application performance
   4. End-user experience
   5. Alerting
2. Understand which pricing model works best for your current infra setup. Each vendor has different pricing models, so try out their free trials and see if their ease of use and transparency in pricing is comfortable for you.
   1. Workload-based pricing
   2. The volume of telemetry ingested
   3. Number of users
   4. Event-based pricing
3. If you’re not quite ready for enterprise volume discounts, look to use multiple tools from different vendors. More than half of the customers surveyed use ten or more tools.
4. Work on your existing strengths. Enterprise customers using these tools have dedicated IT teams set up for monitoring. As a smaller company looking to build on the cloud, leverage automation where possible so you can spend more time on your code and less time watching your systems.
5. Try installing tools like Prometheus and Grafana to understand how much work it is before testing a tool like Datadog. This will help you understand if the time saved using these tools is worth the money you’ll spend on them.
6. Have a way to train your team, and make it easy to onboard new employees into your stack. Once your observability stack is up and running, it is important to have good getting started guides (internal or external), which can help your team learn and work with your observability tools easily.

Conclusion

Choosing between Datadog, Splunk, and New Relic can be difficult as they have similar offerings at first glance. But, there are a few things each platform does better than the others. Using the feature-level comparisons and considerations mentioned above, you can narrow down and make the best decision that saves cost and works well with your stack.

While there is no shortage of observability tools in the market today, choosing from prominent vendors has its advantages in terms of compatibility, cost, and easier monitoring. And that's why companies today are choosing to consolidate their vendors.

If you are interested in seeing the progress the industry has made in the past 12 months and where it is headed, check out this insightful blog by Hayden James.

Argonaut removes the complexity out of your app and infra deployments. You can also integrate third party apps like your favorite observability tool. Get started with Argonaut today.