Modern application delivery leaders realize that their primary goal is to deliver value to the business and its customers faster. Most of the modern successful change frameworks, like Agile (in its various instantiations), Lean, and Lean Startup, which inspire developers and development shops, put metrics and measurement at the center of improvement and feedback loops. The objective of controlling and governing projects to meet vaguely estimated efforts but precisely defined budgets as well as unrealistic deadlines is no longer on the agenda of leading BT organizations.
The new objective of BT organizations is to connect more linearly the work that app dev teams do and the results they produce to deliver business outcomes. In this context, application development and delivery (AD&D) leaders need a new set of metrics that help them monitor and improve the value they deliver, based on feedback from business partners and customers.
So what do these new metrics look like and what can you do with them? In the modern application delivery metrics playbook report “Build The Right Things Better And Faster With Modern Application Delivery Metrics,” I describe:
- Preproduction metrics. Leading organizations capture preproduction data on activities and milestones through productivity metrics, but they place a growing emphasis on the predictability of the continuous delivery pipeline, quality, and value.
- Postproduction metrics. High-performing teams track business metrics to understand how the applications they’ve deployed affect business performance. Typical metrics include application health metrics like deployment success rate, performance stability, and ghost users. But AD&D leaders also track customer-experience-related metrics like clicks, usage, happiness, and preferences.
- How metrics are correlated to identify what’s impacting business. AD&D leaders leverage big data and analytics tools that provide reporting, dashboards, and dedicated algorithms to find correlations (and more) in the sea of overwhelming data generated, which neither the business nor developers could otherwise identify or make good business sense out of.
The use of the above metrics with data and predictive analytics and machine learning makes it possible to discover, for example, poor-quality code that’s causing slow performance for users and consequently poor sales or usage of production features, as well as who is developing this poor code — and it then allows you to take corrective action to fix the problem. Likewise, you can also use data to find out where high-performing teams are significantly contributing to business growth.
To find out more details about these new metrics, the products used, and real examples of organizations leveraging them, read the Forrester report“Build The Right Things Better And Faster With Modern Application Delivery Metrics.”
I am looking forward to your ideas, comments, and suggestions on how metrics can make the world of modern application delivery a better one. I will be kicking off a stream of research on how deep machine learning can make this a better world; stay tuned!