Chronosphere on cloud-native observability insight


It’s a big claim. To say that your firm is the only cloud-native observability platform would be too much for any single cloud insight specialist. For balance then, Chronosphere calls itself the only cloud-native observability platform that ‘tames rampant data growth’ and complexity for engineers. Rampant (in most use cases) being a somewhat subjective measure of infectious contagion and spreading nastiness, the company probably gets away with this term.

Now looking to champion its 2023 Cloud Native Observability Report: Overcoming Cloud Native Complexity, the organization says it has surveyed some 500+ full-time employees in engineering and software development roles who are familiar with observability tools and practices. 

Observability, done right

It suggests that when observability is done right, it positively impacts both the top and bottom lines of a business, allowing organizations to innovate faster, improve customer experience and increase return on investment. 

Conversely, when companies’ observability functions fall short, the results can be catastrophic – amounting to billions of dollars in lost productivity across the US every year. 

The research suggests that engineering and DevOps productivity falters – wasting an average of 10 hours or 25% of their time every week trying to triage and understand incidents. 

Cloud-native demands observability 

As the market increasingly moves toward cloud-native environments, observability must adapt as well if it is to deliver as promised. But really, we’d almost hope and expect Chronosphere to be saying that cloud-native computing is (all about) observability, cloud-native computing demands and necessitates observability and cloud-native computing without observability is a form of myopia, or perhaps even blindness.

The company goes some of the way there i.e. it states that for engineers, not all observability tools are created equal, as companies often choose between vendor’s solutions or building their own. The report found that those using vendor solutions were three times as satisfied, and ultimately, more effective, than those using tools they’ve built. 

  • Those using a vendor solution are detecting issues 65% faster than those without a cohesive approach, and 30% faster than those with an in-house solution
  • 42% of those using a vendor solution said they experienced very severe, customer-facing incidents quarterly or more; much less than the 61% using only their own observability solution who experience incidents quarterly 

Wasting money, talent & time

Few engineers are fully satisfied with their current observability solution, saying it’s too slow, lacks context, and is stifling their performance and ability to contribute meaningfully to the business. In fact, 96% of individual contributors spend most of their time resolving low-level issues but say what they really want to do is innovate. 

“These findings reflect a trend we’ve seen in our customer base for several years – that old school approaches and tools are failing businesses,” said Chronosphere co-founder and CEO, Martin Mao. “During today’s economic uncertainty, companies will gravitate towards solutions that drive engineering efficiency and cost savings while continuing to enable rapid innovation cycles – and this is especially true for companies moving to cloud-native who need to revisit their Observability strategy and solutions.”

Notably, the survey showed that senior engineering leaders are often unaware of the plights of their teams, with more than twice as many individual contributors than senior leaders saying they are frequently stressed (45% of individual contributors vs. 20% of directors). Nearly all of the individual contributors surveyed have complaints about their observability solutions compared with only 12% of senior leaders. 

This ‘corporate digital divide’ is not only costing companies customers and revenue, but also talented engineers who flee organizations because their voices aren’t being heard. 

So once again, yes, we get it… cloud-native computing requires observability first and foremost, front and center, from the outset (baked-in) and for every node and nodule that exists on a modern virtualization-first IT stack deployment. Quite why we can’t get the tech industry to slam it down in those more concrete terms is a still-unfolding story, let’s hope there’s a happy ending.