What Datadog's Acquisition of Metaplane means for the Data Industry

They did it - they finally did it. Datadog for dogs who like data

I remember speaking with some folks over at Datadog corpdev a while back and the line that stuck with me was

“You would not fucking believe how much we get pitched ‘datadog for data’ and it drives us fucking nuts”

Earlier this week Kevin posted

And you have to admire the irony! Datadog finally buying the company that pitches themself as the datadog for data!

I’m impressed some folks finally convinced them. Congrats Kevin, Guru and the rest of the team! We are stoked for you, and I honestly wish I’d been a fly on the wall in those chats esp. RE the DataDog for Data adage…I know you all have said for many years you are the “datadog for data” and it’s awesome the Datadog folks agreed.

I think this is a really interesting acquisition and I’m going to break it down here

No more open-source / data fuckery

There are a lot of acquisitions in the data space around OSS that just don’t really make any sense. You could think of dbt acquiring SDF just to make their compiler actually work or Tobiko’s “acquisition” in response.

There have been a lot other kinda crazy oss acquisitions. The one that springs to mind is Databricks $2bn acquisition of tabular.

But there have been others. Boomi, tangential iPAAS acquired Rivery. Upsolver got bought by Qlik. Snowflake almost bought RedPanda and so RedPanda have raised a massive round instead.

Common to this theme is that everything is either a foundational “tech” or it’s got data movement in there. People are buying tech, or they’re buying compute. SDF is tech. They had no revenue. Redpanda tech is fine, but it’s interesting because of all that juicy data they process.

Rivery and Upsolver, both ELT tools, also processing lots of juicy data.

So where does Metaplane and Datadog fit? Metaplane is observability? There is no compute - it’s pushing queries to your warehouse. Datadog is a tool for Software folks..let’s see..

What do metaplane and datadog do?

I lean on AI slightly here but high-level: Metaplane is one of those companies that fires a load of queries to your warehouse and return the results in dashboards as “data quality tests”. Any engineer worth their salt knows this is a quite frankly terrible line of defence for data pipeline errors.

We should obviously test data as it goes through the pipeline and ideally have it corrected at source. The reason you are willing to pay metaplane is because you’ve created a massive hot expensive mess that now has a big real impact on your business

Metaplane and observability tools in general focus on helping people who are in the “Failure” part of the pyramid

That said Kevin and the team are very smart and they have another feature I really love, which is basically “Blast Radius” — you can identify the impact of your PRs in dbt, code etc. before you make them. Metaplane ingests data from your stack about your lineage and serves this in an API, so you can actually understand the impact of your changes which is really valuable and something folks ask for a lot — this puts you into the $1 portion of the pyramid.

Datadog by contrast is a tool mainly used by software engineers. Software systems don’t normally need orchestration. They’re live, choreographed systems. So to monitor them, you write code that sends logs to something like datadog, a bit like this

import datadog 

def function():
try:
   do_something()
except Exception as e:
   datadog.send_error(e)

Datadog then aggregates all these logs and gives you alerting based on them - you can alert when the number of logs hits a threshold, or below. When certain values get raised, do triage etc.

That said, data engineering is software engineering too! And lots of us data folks use datadog too.

Often, the reason is because our software friends already have datadog and we have python scripts we want to monitor.

What Datadog’s acquisition of Metaplane means

Metaplane is going to carry on as a standalone product. This means existing Metaplane customers are not going to be affected. I don’t think there is going to be any issues for these folks who are Metaplane customers which is nice.

In the medium-term we will see what happens. You would have thought additional cross-selling (“Hey want to buy some datadog”) and price increases would be classic things to come. See here.

Historically datadog have made a fair number of acquisitions. The more interesting thing here is about what metaplane means for datadog customers rather than the other way around.

“CoScreen (2022)
Datadog acquired CoScreen to introduce real-time collaboration features, enabling developers to share screens and control applications simultaneously. This integration facilitates faster incident response and improved remote teamwork, enhancing developer productivity and reducing resolution times. ​Datadog+3investors.datadoghq.com+3Datadog+3

Seekret (2022)
The acquisition of Seekret brought advanced API observability and governance tools to Datadog's platform. This allows customers to automatically discover, monitor, and manage APIs across their environments, improving visibility and control over API operations. ​investors.datadoghq.com

Sqreen (2021)
By acquiring Sqreen, Datadog integrated application security monitoring into its platform. This provides customers with automated protection, threat detection, and security monitoring capabilities, aligning with DevSecOps practices and enhancing application security posture. ​investors.datadoghq.com+1Datadog+1

Logmatic.io (2017)
The acquisition of Logmatic.io added log management to Datadog's offerings, enabling customers to analyze logs alongside metrics and traces within a single platform. This integration simplifies troubleshooting and provides a more cohesive observability experience. ​investors.datadoghq.com

Timber Technologies (Vector) (2021)
Datadog's acquisition of Timber Technologies, the creators of Vector, enhanced its data pipeline capabilities. This allows customers greater control over observability data, supporting on-premises environments and reducing vendor lock-in. ​”

A lot of this tech has just been integrated into the datadog platform. Which is great news for existing datadog customers.

Metaplane has some great features so building these into the Datadog platform are bound to big wins for data folks.

Positioning - expansion or attack?

It’s interesting that Datadog’s method of growth is often to expand its existing customer base. Datadog has high net revenue retention (130%) that it can drive through new products and increased usage of existing ones.

It would feel like an odd move for Datadog to start a full scale attack into new markets, going after the Monte Carlos etc. of the world.

For one, that market is small (observability doesn’t really solve all the problem). For another, what would the value prop be?

Let’s say you have a fairly standard modern data stack. You are bastardising Airflow and running all your data ingestion in it, you have Snowflake, you have dbt, you have Power BI. Everything is great - but stuff is failing all the time.

But now wait! With Airflow, Datadog and Metaplane, all your woes are behind you.

Instead of wondering if your pipelines run properly, you can

• Hire lots of expensive Kubernetes Devs

• Hire lots of expensive Airflow Devs

• Spend lots of time deploying Airflow - note that if Airflow fails to deploy Datadog won’t really help you unless you can get the finicky “heartbeat monitor” to actually ping when a threshold goes below a certain level

• Spend lots of time writing all your logs from Airflow to Datadog — now Airflow is truly just an orchestrator and data processors. It is not used to recover from errors or to look at metadata

• Spend time making everyone get to grips with datadog

• Run data quality tests in metaplane for an additional datadog volume fee, and don’t forget the warehouse costs your incurring by running them too

• Get a nice UI for looking at all your logs and the data quality at rest scores

• Improve your PRs with Metaplane’s lineage detection

• Get another $50k bill for datadog

I just don’t really get what the value prop is here. You’re spending more money, you’re locked into a legacy orchestrator you still have to maintain noone knows how to use or deploy, you have to write all the code yourself which takes time, you have to learn a new tool (datadog), the biggest issue of lineage and rerunning pipelines and making sure your orchestrator actually runs isn’t solved, and you have another vendor and a massive bill!

So in summary - I think this is great news for existing Datadog customers. Who wouldn’t want a bit more firepower?

But it’s a super legacy pattern. The idea you write all your code yourself and do all your observability with datadog and now also use datadog to do all your data observability is nice, but slow, a bit outdated, and fundamentally expensive.

vs. other acquisitions

Datadog is doing something many companies in the data space don’t do. They are bolting on customers and cross-selling the new product. What a crazy and earth shattering idea.

This contrasts with companies buying compute vendors to drive revenue or buying foundational technologies or talent, which seems to happen a lot in the data space.

So it’s quite a boring rationale but it makes sense and is what you’d expect if we were in any other normal industry other than Data and AI!

Summary

It is great to see datadog acquire the data for datadog. Datadog are a publicly listed company so will need to disclose the acquisition price. It leaves other observability companies in an strange place.

There are so many of these companies. Some raised massive rounds, some are still small. Who will buy the small players? What will Monte Carlo do?

We know that observability is a growing market but not as fast as people thought. Companies solve “data downtime” by preventing it with better culture and robust orchestration - not by buying another tool. It has left VCs bitterly disappointed - if DD paid a premium for this people will ask who else will pay a premium? The premium must mean the market is growing fast - shouldn’t we build this functionality to capitalise as well?

I don’t think anyone can afford the companies that have raised massive rounds in this space. They will likely have to consolidate their functionality and laser-focus on smaller, but profitable, enterprise niches and get lean.

The smaller players will probably start looking for exits too depending on the price and how much money has been raised.

We are now in the era where instead of debating what factors lead to what rounds, people look at EV/Revenue multiples and say “why should I pay that much for company X” - key here is revenue. Many of these companies have pretty low revenue and may well be loss-making. For other founders in the space the dynamics of the Metaplane acquisition could be hugely influential as the valuation sets a precedent for their exit opps.

So in summary - as we said before. Consolidation.

---

What is Metaplane?

Metaplane is a data observability platform that helps organizations monitor, detect, and resolve data quality issues in their data pipelines. It’s essentially like a "Datadog for data" — designed to give teams visibility into the health of their data infrastructure, including when and where data breaks, changes, or becomes unreliable.

Metaplane connects to your data stack (think Snowflake, dbt, Fivetran, Looker, etc.) and provides automatic monitoring across several key dimensions:

• Freshness – Are your tables being updated on time?

• Volume – Are there unexpected changes in the number of rows?

• Schema Changes – Did a column disappear or change its type unexpectedly?

• Anomalies – Are there unusual patterns or spikes in your data?

• Null values / Duplicates – Are key fields missing or being duplicated?

If you work in analytics, data engineering, or product teams, Metaplane can help you:

• Catch issues before stakeholders or customers do.

• Reduce time spent on debugging data pipelines.

• Improve trust in data across the company.

Example Use Case

Let’s say you have a dashboard built in Looker that uses a table in Snowflake, and the pipeline updating that table runs through dbt. If the dbt job fails or if a schema change breaks the dashboard, Metaplane can detect the issue and notify your team via Slack before anyone else notices.

How It Works (Simplified Flow)

1. Connect to your data sources (like Snowflake, Redshift, BigQuery).

2. Auto-monitor tables, columns, and metrics.

3. Get alerts when something breaks or looks off.

4. Use their interface to drill down and investigate.

Integrations

Metaplane integrates with:

• Data warehouses: Snowflake, BigQuery, Redshift, etc.

• BI tools: Looker, Metabase, Mode

• Data transformation tools: dbt

• Communication: Slack, email, webhooks

What is Datadog?

Datadog is a cloud-based observability platform designed to help organizations monitor and secure their entire tech stack—from servers and databases to applications, user experiences, and cloud infrastructure. It brings together metrics, logs, traces, and real-time monitoring into a single, unified view. This integration allows teams across development, operations, and security to work from the same data, helping them collaborate more effectively, resolve incidents faster, and improve system reliability.

One of Datadog’s key strengths lies in its wide range of integrations. It connects seamlessly with hundreds of tools and services, including cloud providers like AWS and Azure, container platforms like Kubernetes and Docker, and CI/CD tools like Jenkins and GitHub. Whether you’re running monolithic applications or distributed microservices, Datadog provides real-time insights into performance, usage, and potential issues. Features like application performance monitoring (APM), real user monitoring (RUM), and infrastructure monitoring help teams trace problems from user-facing symptoms all the way back to backend causes.

Security is another critical pillar of the Datadog platform. With built-in security monitoring, teams can detect threats and misconfigurations across their cloud environments and applications. Combined with powerful dashboards, machine learning-driven alerts, and synthetic testing tools, Datadog not only helps teams stay aware of what’s happening—but also enables them to anticipate and prevent issues before they escalate.

In essence, Datadog is like a mission control center for modern software teams. It gives companies visibility across all layers of their technology, allowing them to deliver faster, safer, and more reliable digital experiences. Whether you’re troubleshooting an outage or scaling up your systems, Datadog provides the real-time intelligence you need to stay in control.