Agility Index – Learning loop

• The simple act of monitoring KVAs is a great, but not enough to change the way agility is managed
• The EBMgt approach enables organizations to constantly learn and improve the value derived from software investments
• Its iterative, incremental approach to guided change helps organizations control the risk of disruption, and to compete through their software capabilities
• Learning loop:
• 1. Measure KVMs
  • The first step in the EBMgt learning loop is to establish initial values for the KVMs that contribute directly to the Agility Index™
  • This step provides an initial view of organizational value
  • Representation in a RADAR GRAPH helps visualize relative strengths and weaknesses
• 2. Select KVAs to improve
  • With a clear view of current organizational value and an understanding of the measures that reveal it
  • organizations can now make informed decisions about which KVAs would be most valuable to change
  • Care trying to affect too many KVAs within a single learning loop. Small, incremental changes performed in small learning loops is the most effective
• 3. Conduct Practice Experiments to Improve Value
  • After understanding desired KVAs to improve, select a single or small set of practices to use in an experiment
  • there is practically an endless catalog of practices from which to choose, Scrum.org maintains a list of several hundred generally understood and accepted practices to assist in improving the KVAs
  • Example
    • increase quality to reduce the Defects KVM. A related experiment may be to implement test-first practices in development teams. Making this change in a time-boxed experiment allows observation on the impact of these practices on overall organizational value
• 4. Evaluate Outcome Result
  • The key to understanding results and impact of an experiment is to monitor the trend of value over time
  • Understanding the changes of the KVMs prepares the organization for its next learning loop
  • Remember that the experiment performed in the learning loop is rarely the only factor affecting changes in KVMs during the cycle. Additional factors reflected in changed values may include
    • Marketplace competition
    • Internal policy change, such as new HR processes
    • Release of an update that improves functionality and quality of an existing release
  • Metric values and their affected KVIs change even when the cause is unknown. As these changes are recorded, management can use this data to discuss impact, consequences, and prepare potential responses
  • An advantage for organizations that track changes periodically across time is the opportunity to learn from the patterns that emerge