New advancements in observability mean it’s easier than ever for companies to collect telemetry — but just because you have more data, it doesn’t mean you have more insights. So says Ben Blackmore, CTO of dash0, the new startup on a quest to make observability easy to understand, use, install, integrate, and manage.
Though dash0 launched in 2023 (and announced $9.5M in seed financing in November 2024), the founding team has been working on observability challenges for years. Blackmore, alongside many of dash0’s founding members, worked at Instana, an application performance monitoring (APM) platform, before they sold to IBM in 2020.
According to Blackmore, everyone took a break from observability after the sale, but when they got together just a few years later to see how the space had developed, they noticed a gap. Yes, data collection had notably improved, “[but] we saw [that] a lot of the problems that existed in our time are still problems today,” recalled Blackmore.
It was time to get back in the observability game.
OpenTelemetry May Have Laid the Foundation—but dash0 Is Taking It a StepFurtherr
Leading up to dash0’s launch, the team spent a lot of time looking at OpenTelemetry (OTel), the open source, vendor-neutral collection of telemetry APIs, software development kits (SDKs), and tools for analyzing software performance behavior. In our conversation, Blackmore lauded what OTel is doing to improve data collection for the observability industry: “Previously, you needed a lot of engineering capacity to build data collection. There were few vendors who could do this. But OTel is finally standardizing it. Now you can train employees on the language and tooling of OTel so the majority of knowledge you acquire, you keep.”
OTel is, indeed, a huge win for organizations who want to own data collection and avoid pesky vendor lock-in, a common issue when working with proprietary agents.
Normally, if your organization wants to change to a different data collection solution, you have to start from scratch. “You have to swap out the agent and redo data collection,” explains Blackmore. “That’s a huge pain point — being completely locked into a vendor’s system.”
Dashboards further contribute to the lock-in problem. “There’s a lot of configuration when using observability solutions,” says Blackmore. “With the dashboards you create, there’s a lot of knowledge in there. The problem is, you’re configuring all this for your specific vendor — but then you can’t really take it out again.”
Even solution alerts are subject to vendor lock-in. “When you think about all the alert rules — you don’t want to start from scratch either,” adds Blackmore. “Just imagine if you have hundreds of rules that you’ve tuned over years of usage.”
At the end of the day, no matter how advanced the tooling is, proprietary data collection agents stifle observability teams by making it difficult for them to truly own their institutional knowledge. That’s the beauty of open source — and why dash0 is leading with OTel.
Richer Metadata — Without the Price Tag
In fact, Blackmore says dash0 is such a proponent of open source that they decided to build their solution on top of OTel.
dash0 is a landmark OpenTelemetry-native observability tool that eliminates vendor lock-in and makes it easier to collect relevant telemetry at lower costs. While you’ll find other tools on the market today that also claim to be OpenTelemetry-native, dash0 does more than just free you from vendor lock-in; it extracts meaningful insights from your data to speed up and improve problem-solving: “We turn the raw data into information,” says Blackmore. “It’s not just about presenting the data to you but allowing you to truly use it.”
Making data “usable,” starts with enriching metadata, i.e., the context that makes logs, traces, and metrics actionable. But getting that metadata is often prohibitively expensive.
As Blackmore explains, “Let’s say you have a specific app or service and you want to annotate it. With the majority of solutions, you’re paying per gig—not per log record.” This pricing model essentially de-incentivizes organizations from enriching data—or at least, makes it financially burdensome.
dash0 is changing the tides with a different, more affordable pricing model.
“We believe in rich metadata,” says Blackmore, “which is why we have chosen to price by things that every developer can just count: log record.”
How does that help? It dramatically simplifies cost attribution and cost optimization. Your costs still scale with usage, but now it’s a number you can track and control. For example, “if you want to know which service is generating the most cost,” Blackmore goes on, “just count how many logs have been recorded by service.”
Log Analysis — With AI Assistance
Still, enriching data is only one part of the equation. Once data is collected and enriched, many teams struggle to make sense of their data so they can truly use it.
According to Blackmore, this is a classic problem — and up until now, there hasn’t been a good way to fix it.
When teams collect log records, they’re unstructured and difficult to interpret. “You might have text, but you might now know if it’s an error to investigate or not,” he says. A log might indicate a failure, but there’s no real way to know if it’s truly a critical issue or just routine behavior. Ultimately, this means engineers have to waste time manually sifting through logs to try to piece together relevant details.
Just as they’re making data enrichment easier and more cost-effective, dash0 is making error detection and log analysis more intuitive. And they’re using AI to do it.
Specifically, dash0 AI-enhanced approach automatically parses logs, allowing teams to immediately see what’s causing trouble — no manual digging required. “When you filter for errors, the relevant logs come up right away,” Blackmore explains. “It also shows the distribution of this product ID, as part of our new triage feature.” In other words, with dash0 Triage, you can analyze error distribution across different attributes, like specific product IDs or user segments. This makes it easier to identify the most relevant data — without having to do the manual work to get there.
dash0 Provides the Context To Make Telemetry Actionable, Faster
Bigger picture, Blackmore stresses the larger focus of dash0 is to provide context to help organizations better navigate, understand, and use telemetry: “Any time we show you a piece of data, we’re trying to put it into context.”
For example, dash0 helps teams quickly identify the most relevant attributes within their telemetry. With most other solutions, all data is treated equally — but Blackmore says this isn’t actually reflective of reality: “Some information is more important than other information.”
dash0 helps you make the distinction, interpreting your data, prioritizing the attributes that matter, and surfacing only the important insights you need to problem-solve. This way, you can skip searching through an endless list of metadata and get to troubleshooting faster.
The observability tool also provides context to help teams make better sense of their metrics.
“When you’re working with metrics, you want to know more than just, ‘I have this metric and it’s costing me this much.’ You also need to know where it’s coming from and through which mechanisms it’s been collected,” explains Blackmore.
That’s where he says dash0’s Spam Filters can help. Instead of constantly filtering out irrelevant logs after the fact, you can use Spam Filters to stop noisy data at the source. Specifically, the point-and-click feature lets you identify and block unwanted telemetry data so that only relevant, actionable data gets stored.
“It’s very easy to collect stuff you don’t care about,” Blackmore pointed out. “With Spam Filter, you don’t just throw it away; you stop it from being collected in the first place.”
Bringing Context and Clarity to Telemetry Confusion
From data collection and enrichment to log analysis, dash0 provides the context teams need to turn data into actually actionable information they can use to problem-solve—faster and with lower costs.
When it comes to observability, more data can sometimes just mean more confusion. dash0, it seems, is providing some much-needed clarity.
The post With OTel, dash0 Wants to Make Observability Actually Useful appeared first on The New Stack.