We sat down with Steve McCanne, CEO of Jut, a new platform that provides real-time insights for dozens of business data sources, on the day the company opened its beta product to the public.
How does the proliferation of software change the nature of business?
Whether you’re a startup, Facebook or General Electric, software is becoming the core of the business for innovative companies in every industry. We believe that ‘DevOps’ and developers play a key role in building the next generation of businesses because they are tasked with understanding how software performs, how users are interacting this technology, where the bottlenecks lie and where it can perform better.
The question is, how can companies better “program” analytics and visualizations into their infrastructure to make sure software is functioning as intended? When architected properly, the DevOps team plays a key role in the process by implementing a data platform and – in turn – creating better understanding among developers, operations, product managers, data scientists and business analysts.
Jut is opening its beta this month – how does it play into this mega-trend of software adoption?
The approach to building software and instrumenting it has been fragmented and inefficient. There are two key players – builders and buyers. The builders tap into open source projects like Spark, Hadoop and Elastic Search and build their own applications on top of them. Buyers comprise of DevOps teams with an array of tools to better understand the technological environment. With numerous silo’d tools, it is almost impossible to correlate data like revenue with system errors, or usage patterns with new code. In both cases, you’re stuck with a mess to maintain and upgrade.
Jut's approach is pretty different. We're an operations data hub – a place where any and all operational data (metrics, logs, and system events) is aggregated and organized in a single platform so that any question can be asked about the data by anyone, at anytime. We’ve built an analytics and visualization layer to stream analytics in an agile and ad hoc fashion. We do this without forcing you to use a low-level system language like Java or Scala.
Our open beta offers this functionality to any company to try this approach at any scale, at no cost.
With a more prolific amount of data, businesses will need better tools to make sense of their information. What are the characteristics of the next-generation data platforms?
The world needs smarter approaches to collect and analyze a huge amount of data – the answer is not to create a tool that works with small, siloed data sets.
Instead, data platforms must recognize that not all data is the same, and businesses will have unique needs for their data. Technology like Jut must offer readily accessible data via a platform that’s flexible enough to let developers analyze the data that matters to them - at scale and in real-time. Developers and DevOps need to know how their software is performing now, not 15 minutes ago, not yesterday.
What challenges do you face as a startup in this industry?
As powerful as our platform might be, it’s challenging to sound different. We’re seeing rapid innovation from numerous players, which both keeps us on our toes and reminds us that we have to constantly innovate and offer a truly unique, valuable product.
My advice is to focus, focus, focus. Focus on a particular buyer and his or her problems. There are many smart, experienced people who will help your startup if they really understand how your technology will solve pressing problems. By focusing on that user, you'll be able to scope your MVP and more effectively create a relevant message.
But, you have to build for the user while architecting for their entire business. Build a platform, not just a tool – stay focused on continuously adding value to your customers.