Workload Automation Blog DevOps Blog

4 Essential Takeaways from the 2017 State of DevOps Report

Tina Sturgis
by Tina Sturgis
3 minute read

It’s no surprise that embracing the DevOps philosophy is becoming increasingly critical to the success of digital business. The 2017 State of DevOps Report,1 based on a worldwide survey of 3,200 people, reinforced the growing importance of automation and its connection to achieving positive business outcomes. Additional key findings focused on continuous delivery and how all DevOps teams — regardless of where they are in their DevOps journey — are improving their deployment frequency and lead time for changes over last year’s results. Plus, developers are becoming more empowered to innovate and drive business growth.

Here’s a closer look at some of the findings and how they reinforce the need for taking a Jobs-as-Code approach, where developers can code automation directly into the pipeline from the start.

1. Automation frees up organizations to innovate

When developers are forced to rely on too many manual functions, they tend to get bogged down in rework for testing, deployments, configuration management, and change approval processes. The high-performing DevOps teams surveyed are spending a lot less time on manual efforts compared to what the report refers to as low and medium-performing DevOps organizations. (The report uses a cluster analysis based on metrics for deployment frequency, lead team for changes, MTTR, and change failure rate to determine the IT performance levels). Here are some examples based on the report:

Percentage of Work Done Manually Based on DevOps Performance Levels2

Function High Performers Medium and Low Performers
Configuration management 28% >45%
Testing 35% Approximately 50%
Deployments 26% >42%
Change approval processes 48% 67% for medium performers and 59% for low performances. This may be because medium performers have automated enough work for desired results but have used the freed-up time from automation to add more manual controls around changes.

By shifting the change review process to an earlier phase of the development cycle — think “shift left”— manual work can be reduced and automation increased. The Control-M Automation API, for example, enables developers to take a Jobs-as-Code approach with the autonomy to code, debug, and test workflows early on so they can deliver applications faster, more efficiently, and accurately while improving batch processing. Actions needed to execute a job become part of the code.

2. A Little DevOps Goes a Long Way

While the high-performing DevOps organizations are doing better than the low and medium performers in terms of throughput and stability, the low-performing groups also achieved some significant improvements over the prior year. For example, they are shipping code more frequently and reducing the lead time for changes.

Other report highlights include:

  • High IT performers deploy code on demand with multiple deployments per day. Low and medium performers deploy between once a week and once a month.
  • High IT performers require less than one hour lead time for changes. Low and medium performers require between one week and one month.
  • High and medium performers have a change failure rate of 0-15% while low performers have a rate of 31-45%.
  • The mean time to recover for high performers is less than one hour. For medium performers, it’s less than one day. It’s between a day and a week for low IT performers.

3. Focus on Continuous Delivery – Deployment, Automation and Automated Testing

The report evaluated the effectiveness of continuous delivery based on a team’s ability to deploy on demand to production or to end users throughout the software delivery lifecycle. This evaluation also included the ability of development to provide and act quickly upon fast feedback on the quality of a system and ease of deployment.

The report described how automation contributes to continuous delivery. With automation, organizations should be able to do most of their testing on demand. The Control-M Automation API, for example, allows developers to find defects and bugs early in the delivery lifecycle. This lowers costs, reduces manual rework, and improves application quality.

4. Empower Development Teams

The report explained why developers should seek input from customers throughout the delivery process and why developers need the flexibility to change requirements based on what they discover. They should be encouraged to try out new ideas to increase the likelihood that what they build will drive value.

To meet business demands, developers need digital business automation to work quickly, effectively, and collaboratively. The Control-M Automation API streamlines application release cycles by operationalizing applications faster, improving quality and increasing staff productivity. “We have been using Control-M for years in operations and now the product gives our developers full ownership and control of their jobs in a coding environment that is familiar to them, so they can define the business processes they want to automate in production,” said Robert Stinnett, Automation Analyst, IT Operations at Carfax.

Read the ebook below to learn more about how to DevOps can deliver higher-quality applications faster, easier, and more effectively.

1 2017 State of DevOps Report, presented by Puppet and DORA

2 2017 State of DevOps Report, presented by Puppet and DORA

Build continuous delivery into your organization

Operationalize applications faster with automated job scheduling
View the E-book ›

These postings are my own and do not necessarily represent BMC's position, strategies, or opinion.

See an error or have a suggestion? Please let us know by emailing

About the author

Tina Sturgis

Tina Sturgis

Tina Sturgis is currently a Solutions Marketing Director focusing on Control-M Automation API and bringing more Ops into Dev with Jobs-as-Code. With her she brings 20+ years' enterprise software sales, services and marketing experience. Prior to joining BMC in September 2016, Tina spent nearly 13 years at Hewlett Packard Enterprise Software focusing her expertise in project and portfolio management, application lifecycle management and most recently DevOps across the HPE products and services portfolio. Her vast DevOps experience lies in how to implement a DevOps methodology, what a DevOps operating model should look like and how to make DevOps really work inside complex organizations by measuring success and focusing on organizational change. She earned her BBA degree in accounting and economics from the University of Michigan.