DEVOPSdigest

Software is at the center of every business, increasingly becoming a strategic component of offensive and defensive business plans and strategies. Companies' digital transformation initiatives, and short- and long-term economic recovery plans, can't be accomplished without tactical and strategic pivots, which are most often reliant on software. To pivot successfully requires investments in software and the teams who create that software.

For business leaders to justify their software investments, though, software teams in product organizations, in IT and externally must have a track record of quickly and reliably delivering mission-critical capabilities and features important to the business and its customers. Software must deliver the expected value in the market and internally, with the ability to course-correct based on learnings and results. Leaders must have confidence in their software organizations' ability to experiment, fail, recover from failure and then pivot quickly to deliver needed results. Software Delivery Management (SDM) provides a framework, strategy and discipline to benchmark progress, track and manage resources and ultimately deliver results.

July 21, 2020 saw the release of Accelerated Strategies Group research report on The State of Software Delivery Management 2020 commissioned by CloudBees.

ASG's research delved into serious questions about organizations' ability to successfully manage multiple, interdependent and complex software delivery pipelines, provide insight and visibility into the state of workproducts, assess the impact of shifts required by the business, evaluate the cost of delivery failures and assess the fidelity of communications within and across the organization.

The report includes a formal definition of SDM. Simply stated, SDM is about how well we manage software creation and delivery across technical and business disciplines, sharing end-to-end information with fidelity vertically and horizontally within the organization, to ensure the expected outcomes: timely, impactful business value.

The research shows software organizations making good progress toward achieving these outcomes, as software teams continue their shift away from the traditional IT “order taker” role of the past to new norms where they're delivering value. In fact, according to the research, 67% of organizations said they are able to prioritize development features based on the expected business impact. SDM better enables organizations to know upfront the expected business impact that comes from investing in creating new software, improving software capabilities or adding new features. In the research, 61% of respondents said SDM shortened lead times for feature delivery, which is one of the key metrics for measuring an organization's software delivery maturity. While both results are relatively good news, there is still substantial work to do.

While software organizations feel SDM helps them understand the expected business impact of new software, capabilities and features, it's still difficult to measure opportunity costs and negative consequences to the business if those aren't implemented. More than half of respondents could not quantify the impact of feature delays or the cost of defects found post-delivery.

Other improvements the research seemed to suggest center around team and organizational communication and collaboration, as well as how data and information is accessed and shared. Organizational silos are still prevalent in organizations, with more than 85% of respondents indicating silos impact the free flow of information across the organization.

Organizational silos, entrenched behaviors that restrict information flow and hamper collaboration between vertical organizational structures and functional responsibilities within the company remain a major obstacle. Silos within software organizations and between levels of hierarchy in the business, lead to senior managers, product owners and DevOps teams each having a different view of delivery progress than those closest to software development. Common behaviors like restricting access to information or hoarding institutional knowledge and limiting collaboration put the successful adoption of DevOps and Agile at significant risk of failure.

Automating highly repetitive processes like continuous integration/continuous deliveryt (CI/CD), testing and deployment generate a great deal of untapped data with insight into the progress, efficiency and areas for improvement across the software delivery process. Despite the increased use of integrated tool chains and automation for testing, development environment setup and software deployment, most respondents indicated they do not have access to good quality data from across the tool chain. This was also true about data across multiple applications and projects.

ASG believes this lack of access to data is an indicator of deeper issues. The software delivery process would benefit from greater efforts to identify existing data in the toolchain, as that data may be beneficial for measurement, metrics, goals and improving value delivered to the business. While easier said than done, there may be existing data that, when surfaced and rendered useable, would solve some of the issues related to access to data, or lack thereof.

More in-depth information, statistics and a fuller explanation of our findings are available in the full report. As implied by the year 2020 in the title, ASG will continue this research in the future.