Traditional DevOps initiatives have enabled businesses to become more Agile, automate more, make smaller changes, and get releases out faster… but that’s not enough to compete in today’s modern development world.
According to Mike O’Rourke, chief research and development officer at Digital.ai, there is a missing link between DevOps and the business. “There is no link between the things the development organization is doing in the DevOps space in terms of the way they are managing their success versus the way the business is managing their success,” he explained.
Success tends to differ depending on whose measuring it. For instance, success for the development team might be how fast they are able to respond to changes, get a release out or how many bugs they can detect and fix before a release goes out. For the business, success might be getting a 5-star app rating or improving customer satisfaction. Both sides matter, but the business and development teams really need a way to understand whether or not what they are measuring is having positive or negative outcomes.
This is where the notion of value stream is becoming extremely important to extend the scope of DevOps and achieve business success. “As you go through and you are developing something, wouldn’t it be great to know what the business expects to get out of that?” O’Rourke said. “In most companies out there, there is a big gap between that understanding. What are we trying to do as an organization is dramatically reduce the gap to the point where when the business has an outcome they are looking for, everyone in the development organization understands what it is they are looking for, why they are looking for it, and how they can help.”
The first step in getting there is creating a value stream map, according to O’Rourke. A value stream map maps each individual piece of the process and links it to the entire development organization so teams can identify potential places to improve and add more value.
“The key is to be able to flow information throughout the development life cycle. That is something that DevOps doesn’t necessarily do today out of the box,” said O’Rourke. “What DevOps does very well is it looks at things like burndown charts and how many releases you did, where you build failed, etc. etc. But actually watching the value flow throughout the development life cycle is something that they are missing.”
It’s also not just about flowing information throughout the pipeline. It’s about being smarter historically about changes that occurred, how they are linked, and what was the result. Traditionally, the DevOps process has stopped when an item has been delivered. Value stream takes it further and enables teams to gather information about how customers are using an application, which features they are using, why, how it is performing under load, and how secure it is. “All this development information now gives me a much richer set of insights that I can hand back to the business about what is really happening with their app and because I know why they were asking me to do this work, I can present it in ways that are interesting to them,” said O’Rourke.
Digital.ai helps DevOps teams expand their reach and become a value stream management organization by having context management built right into their platform so everyone can understand what tasks and defects are related to a certain set of checks, builds, pipelines or releases. “This notion of value stream mapping now flows throughout the whole system and as long as you are using some tool whether you are doing OKRs or value stream mapping, as long as you can flow that in to the beginning of the pipeline, we will make sure that flows throughout the rest of the pipeline,” said O’Rourke.
The company also integrates with popular application performance monitoring, AIOps and ITSM tools so it can gather information regardless of the environment the app is running in. When an app gets deployed, it can collect everything from the operating system to the memory size and bring all that information about what is happening inside the application back to the developer and business.
For organizations using their own DevOps tools, Digital.ai provides an artificial intelligence add-on, which includes machine learning and analytics built into a conformed data lake.
“We do believe everyone is going to want to go to value stream management overtime, but most organizations are still struggling just trying to get this information back to their own teams,” said O’Rourke.
Learn more at digital.ai.
Content provided by SD Times and Digital.ai