This year, more than 4,600 technical IT and DevOps professionals from around the world shared their experiences in the Puppet Lab’s 2016 State of DevOps Report, helping the industry deepen its understanding of the technical practices, cultural norms and lean management practices that make up DevOps. What’s more, having this type of information helps other teams using DevOps practices benchmark their performance compared with others. Learn more about DevOps and see how your team measures up.
DevOps has caused a lot of questions in the IT world lately. If you have heard this term floating around, you may have thought to yourself: what exactly is DevOps? Or, if your team already has a handle on it, you may have wondered how to measure its success. Is it the pace of continuous delivery? Or how often your team can push changes with limited failures? How does one measure the efficacy of integrating teams?
Well, now you can. This year, more than 4,600 technical IT and DevOps professionals from around the world shared their experiences in Puppet Lab’s 2016 State of DevOps Report, helping the industry deepen its understanding of the technical practices, cultural norms, and lean management practices that make up DevOps. What’s more, having this type of information helps other teams using DevOps practices benchmark their performance compared with others.
So, read on to explore the 
Agile deployment practices
Companies like Etsy deploy up to 80 times a day, and Netflix and Amazon deploy up to thousands of times a day. This is the heart of agile deployment practices: enabling developers to practice early delivery and continuous improvement by shipping frequently.
As a result, their lead times (i.e. the time it takes for code committed to be deployed in production) have narrowed down to less than an hour. Moving fast allows these companies to stay on the cutting edge of the industry.
There are ways your team can achieve this, too:
- Lean product management: Break up major releases or changes into smaller chunks that can be released over time.
- Branch to trunk development: Change from feature branches to trunk based development (or working off the master code with version control instead of feature branches). Hint: Bitbucket can help with this!
- Waterfall to agile: Move from a “waterfall” software development process to an agile development process
- Automation: Invest in automating processes after committing code. Bamboo, Bitbucket Pipelines, and Docker are great tools for this.
The report has found that teams that practice agile development deploy 200x more frequently than low performers, with 2,555x faster lead times.
Now that’s fast. “Teams that practice agile development have 2,555x faster lead times”
Integrated teams
It doesn’t end there. You may have heard of the principle of the three ways in DevOps by now, but essentially they all sum up to one basic premise: integrating teams across a workflow. Information technology (IT), Developers (Dev), Information Security (InfoSec) and Operations (Ops) teams accomplish this by sharing visibility and opening up feedback loops.
For example, when InfoSec and IT are better plugged into the entire workflow, developers are considering potential architectural flaws while writing code, as well as incorporating security requirements within their automated testing – things that otherwise would have been found downstream by InfoSec late in the process, or even have caused incidents.
How can you integrate your teams? Some ideas:
- Think about process: Include other teams within your daily standups, as well as demos.
- Share a common set of tools: If IT and Ops teams use Jira Service Desk, Hipchat, and other tools to collect incidents and monitor performance, they can integrate tickets within Jira Software tickets.
- Incorporate requirements within automated testing: Collect requirements from IT, InfoSec, and Ops teams and incorporate those within automated testing.
What happens when teams integrate? According to this report, they recover 24x times faster and have 3x lower change failure rates. To top it off, they spend 50% less time remediating security issues.
Ops teams, rejoice! “Integrated teams recover from incidents 24x faster than the rest”
Everyone is an owner
As a result of everyone being integrated, organizations can experience a cultural change: everyone becomes an owner. Everyone is responsible for the end result that customers experience.
Devs are more incentivized to write quality code that will not cause incidents for IT, Ops, and InfoSec teams. IT, Ops, and InfoSec teams feel more responsibility to close the loop between incident and resolution.
Teams that experience more ownership over the product and code spend 22% less time on unplanned work and rework. As a result, they have more time for new work: designing and building new features.
This means that organizations have a better shot at creating superior products and experiences – which translates into business value and more revenue from appreciative customers!
Ownership for the win. “Teams that share ownership spend 22% less time on unplanned work and rework”
How does DevOps compare to agile?
You may at one point scratched your head and asked yourself…how does DevOps compare to lean or agile? 
Essentially, it is agile development practices combined with tight integration with operations, all of which are based on lean principles. This visual may help, pulled from the Agile Web Operations blog:
DevOps is a revolution that is happening right now. Over the past two years, throughput by teams practicing DevOps has increased by 630%, and change lead time has reduced from days to minutes over last year’s top performers.
Show me the numbers
You read through what DevOps is. You understand that’s a big deal, and have read about how high performing teams are benefiting from it. If you want to benchmark your team’s performance, check out this table from the 2016 State of DevOps report.
So, how does your team compare?
Excited to get to work?
Read through the full report here. And, read through our DevOps ebook for more about how Atlassian does DevOps in product management, software deployment, and incident management processes.