I’ve been reading the DORA State of DevOps 2018 report, and it does make for an interesting read.

The report uses cluster analysis to benchmark organisation against the collective industry.

This is a good report, but like others, the report leaves much for the reader to consider. How does my organisation fit against the benchmarks? Does being an elite or high-performer by DevOps standards yield in subsequent gains to the bottom-line of my organisation?  My organisation is already amongst the industry leaders, are these capabilities I need to exhibit?

Organisations reading the report should consider the real value of making multiple deployments in a single day or having a one hour turnaround time for changes in production against what matters most to them, e.g. growth and profit, before setting sights on becoming an elite performer.

The content of what is being deployed should be more important than achieving multiple deployments in a day.  However, few organisations need to make multiple deployments a day to keep the pipeline of feature releases and updates flowing, these organisations will typically have a modern microservices based architectural landscape spanning hundreds of applications and services.  These organisations are more likely to be classified as high or elite performers, anything less, profitability markedly suffers.  Does this sound like your organisation?

The report also provides an interesting view on outsourcing and its impact on software delivery performance.  Nothing new, DevOps was conceived in an outsourcing world of silos, friction and ‘handoff’ waste.   Needless to say, the report correlates outsourcing as a key trend in the organisations ranked as low performers (from a DevOps perspective).  It then uses this trend to explore the concept of the financial ‘cost of delay’ using Maersk as a case study.  The report is seemingly trying to convince those organisations who primarily adopt outsourcing models to save costs to re-think their strategy.

The report introduces consumers to a distinct new term in its own right, observability. It seems the managed-service-providers and value-added-reseller can distinguish the differences between monitoring and observability, everyone else needs to catch up and add observability to their list of must-have capabilities.  The report provides clear definitions for monitoring and observability.

Culture has always been a mainstay of the DevOps movement, the report takes a different view on the relationship between DevOps culture and performance.  Where in the past, DevOps culture was mainly discussed as a definition, i.e. the end-state, this report focuses more on how to influence culture as a leader within your organisation.