How can DevOps enable your agile transformation? Insights from the State of DevOps 2017

Organisations today are looking to significantly speed up delivery and adopt an ‘always be releasing’ and continuous delivery mindset. This is in response to the pressures that are increasing from customers who are demanding quicker turnaround of new or updated products/services and/or new entrants that are creating competitive pressures on large organisations as they are able to innovate faster.

To achieve the ‘always be releasing’ mindset, organisations will need to look end-to-end and transform to re-invent capabilities across both the business and IT to evolve towards an agile organisation. A critical foundation for “moving fast without breaking things” is building an Agile Engineering and DevOps organisation.

Today, DevOps is an understood set of practices and cultural values that has been proven to help organisations of all sizes improve their software release cycles, software quality, security, and ability to get rapid feedback on product development

Summary of the 2017 State of DevOps Report

For the last number of years, the State of DevOps report has collated some of the most pre-eminent thought leadership and research about the adoption and success of DevOps in organisations.

Deloitte is proud to be a sponsor of the 2017 edition and recognises that a mature DevOps capability is a key enabler to helping shape and transform the agile enterprise.

This blog will explore 4 of the key themes from the State of DevOps report 2017 and examine common challenges based on experience from the Deloitte front-line.

1. Transformational leadership enabling the success of your teams

It is now widely accepted that cross-functional, co-located and diverse teams which collaborate together get the best outcomes. As the root of DevOps has typically stemmed from building exceptional bottom-up technical practices, the impact of effective Leadership in software delivery can sometimes be ignored.

Characteristics of transformational leadership are highly correlated with IT performance

It is significant to have senior leadership that is supportive and buys into the organisation’s direction and how IT can be a key driver to achieving the business objectives.

It is then in equal parts the role of a ‘Transformational Leader’ to enable their high-performing teams to deliver great work as well as create an environment with practices and standards that breed technical excellence. The report describes the 5 dimensions and characteristics of the Transformational leader – vision, inspirational communication, intellectual stimulation, supportive leadership, and personal recognition.

There is also a subtle, but key difference between transformational leadership and servant leadership. As a transformational leader there is a greater focus on outward and upward delivery towards strategic objectives, versus a servant leader’s focus is on protecting and enabling the team.

Good leaders help build great teams, great technology, and great organizations indirectly, by enabling teams to re-architect their systems and implement continuous delivery and lean management practices.

The questions that ‘Low Performers’ need to ask themselves is:

  • ‘Do you have the right transformational leadership in place?’, and
  • ‘Is change being driven top-down – or is it still expected to organically grow through the grass-roots?’

2. Build quality in, deploy early and often, automate what you can

The relentless focus on businesses being quicker and better hasn’t changed. This is evident in the narrowing gap over the last 12 months between how frequently low and high achieving IT performers deploy to production.

Low performers went from shipping between once per month and once every six months in 2016, to shipping between once per week and once per month in 2017.

However, the report points out that the stability of the high performers has improved, demonstrating that there is a greater ability to recover from incidents and proactively prevent them in the first place.

Adoption of test automation is most likely a leading indicator for positive IT performance. These same organisations are able to automate configuration management, testing and deployments by ~25% more on average than the lower performers.

High performers understand that they don’t have to trade speed for stability or vice versa, because by building quality in, they get both.

Our experience shows that oftentimes ‘going agile’ creates the myth or misconception of ‘going fast at the expense of quality and stability.’ Automation has been a focus for low-performing organisations as a first step, but there is still more work to do to uplift throughput metrics and ‘shift left’ test and quality controls to minimise the technical debt build-up and cost of finding quality issues late in the delivery cycle.

3. Loosely coupled architecture and independent teams

Just as we promote teams to operate autonomously, having a loosely coupled architecture allows for management of dependencies to be internal within a team and limits the amount of work that needs to be completed on outside platforms.

This enables more frequent deployments and reduces the inherent technical complexity when delivering functionality and integration with multiple systems and data sources.

Organisations should architect around team boundaries to ensure that teams can get their work done — from design through to deployment — without requiring high-bandwidth communication between teams

This may also represent a significant cultural or maturity change for many organisations in the way they are comfortable operating today. It may also imply that the team has capability and maturity to:

  • Allow the teams to use whatever tools are best for the job, rather than being mandated by a central function
  • Make large changes without outside permission and during business hours with minimal downtime
  • Test outside of an integrated environment
  • Deploy independently of other applications, on demand, when ready

4. Lean product management, experimentation and a focus on the customer

IT performance predicts lean product management practices. Improving your software delivery pipeline will improve your ability to work in small batches and incorporate customer feedback along the way.

In an environment that is able to release working software on demand, there is less need to deliver projects or initiatives of significant size, as it is realistic to deliver valuable work incrementally. This is not the case in many larger organisations which have longer lead times before they can drop functionality.

Not only do smaller batch sizes aid the speed at which change can be delivered, they reduce complexity and risk of delivery. Short lead times also enable the business to be adaptive to feedback and directly focus on what the end customer wants.

This tighter feedback loop also means that the appetite to experiment and test new ideas is often greater, making techniques such as A/B testing (publishing two versions of the same product to see which works better) more effective. To understand how to create an ‘experiment friendly’ culture, please refer to our earlier blog post.

Summary

Embracing DevOps and new ways of working is not just about building leading technological practices, but also shifting the dial on cultural and leadership throughout IT and the organisation.

Businesses which are able to be adaptive to external factors and build a stable yet dynamic technology ecosystem will continue to gain a competitive market advantage over the slow-movers, or those resistant to change.

 

References

All references are sourced from the 2017 State of DevOps Report.

Link to the full 2017 report: https://puppet.com/resources/whitepaper/state-of-devops-report


Want to stay up-to-date?

Stay on trend and in the know when you sign up for our latest content

Subscribe