Consumer demand for quality and speed of digital experiences has reached unprecedented levels, yet the cloud-native technology underpinning these experiences remains difficult to manage and an inhibitor to progress.
In response, many Australian organisations are establishing platform engineering functions and tools to aid in digital experience delivery.
This is very much on-trend. Analyst firm Gartner recently identified platform engineering as one of the top five strategic technology trends in software engineering for 2024, predicting “that by 2026, 80% of large software engineering organisations will establish platform engineering teams, up from 45% in 2022.”
Platform engineering teams typically oversee the creation of centrally-managed internal developer platforms (IDPs) that give the various product and digital delivery teams inside of an organisation the components, tools, and templates they need to get on with creating solutions, while abstracting away the management complexity.
Automation is key to that abstraction. It’s used to help teams manage their infrastructure and workflows, eliminating extraneous manual processes and enabling teams to develop, test, deliver, deploy, and execute other key processes at scale.
But not every platform engineering team folding automation into IDPs - nor every team using the automated features in the IDPs - has achieved a targeted level of sophistication in what it’s able to do.
Recent research by Dynatrace found that an average of 56% of end-to-end DevOps processes are automated across organisations of all kinds. Beyond the average, maturity levels vary widely, with every organisation at a different stage in its journey based on a multitude of factors, from the level of executive buy-in to the resources that are available and the culture for change that exists within teams.
Once at a mature level, organisations with automated workflows, repeatable tasks, and other DevOps processes can not only exponentially accelerate software delivery and business growth but also improve employee satisfaction and productivity.
An assessment approach
The first step toward achieving these benefits is understanding exactly where an organisation’s current automation maturity level stands. This knowledge empowers teams to embrace an informed and systematic approach to improve maturity.
One way of doing this is by completing a DevOps Automation Assessment. This involves assessing maturity levels across four areas to build up a nuanced understanding of where an organisation’s automation maturity stands. Since teams from one functional area to another may be siloed, a respondent who is not knowledgeable on the automation practices of a certain area can still obtain insights by answering the questions that pertain to their team’s responsibilities.
The first aspect considered is an assessment of automation governance - the extent to which an organisation prioritises automation efforts, including budgets, ROI models, standardised best practices, and more. The next assessment should be of development and delivery automation - the extent to which an organisation automates processes within the software development lifecycle (SDLC), including deployment strategies, configuration approaches, and more. From there, operations automation should be assessed to determine the level of automation used in maintaining and managing existing software. Finally, an assessment of security automation should be used to verify how much automation an organisation uses when mitigating vulnerabilities and threats.
From this, an overall score can be calculated to determine and display an organisation’s maturity level.
The four ‘grades’ of maturity
The most basic level of maturity to be attained is ‘foundational’. At this level, automation practices are either non-existent or elementary and not adding significant value to DevOps practices. Organisations at this maturity level should aim to build a strong automation foundation, define automation principles, and lay the groundwork for a more mature automation framework.
The next maturity level up is ‘standardised’, denoting situations where automation has become more integrated into key DevOps processes. This includes expediting workflows, ensuring consistency, and reducing manual effort somewhat.
Above that is an ‘advanced’ maturity level. At this stage, automation practices are integrated across the software development lifecycle and assist greatly in scaling and executing DevOps processes.
Organisations at this maturity level should strive to improve operational excellence by adopting AI analysis into their automation-driven practices to reach the highest maturity level - ‘intelligent’. At this level, organisations are leveraging artificial intelligence and machine learning (AI/ML) to bolster their automation practices.
Even having attained this top maturity level, the goal of organisations should be to achieve even higher levels of efficiency, agility, and innovation. Teams can advance workflow automation further by adopting unified observability and log data collection, along with different forms of AI (predictive, causal, generative) for automated analysis.
A roadmap of meaningful improvements
While it’s helpful to understand maturity levels, the information is only useful if teams can leverage it for improvement.
For example, to progress from standardised to advanced, it is recommended that organisations implement a single source of reliable observability data to prioritise alerts continuously and automatically.
Or, to progress from advanced to intelligent, the report encourages organisations to introduce AI/ML to assist continuous and automatic security processes, including vulnerability detection, investigation, assignments, remediation verification, and alert prioritisation.
Even at the highest level of automation maturity, there is always room to continually improve automation practices and capabilities as underlying technologies and the teams that use them evolve.
