Improved metrics to demonstrate the success of agile methodsJason Tice
schedule 1 year agoSold Out!
Each year VersionOne conducts a survey to assess the state of the agile community, highlight emerging trends, and provide insights on practices and techniques that are proving successful. I observed an interesting and alarming trend within the 2015 survey data. Many in the agile community have commented that they frequently are asked to evaluate the effectiveness of agile methods, which leads to questions about how do we measure success. According to VersionOne’s 2015 survey data, the most frequently tracked agile metric to determine success on a daily basis is “Velocity” - Jim Benson (author of “Personal Kanban”) has referred to velocity as “an imaginary number divided by an arbitrary amount of work” - it is a very imprecise metric due to the high variability and non-scientific origin of the data used to compute it. I hypothesize that many of the questions surrounding the effectiveness of agile methods result from focusing too much on the use of imprecise metrics, such as velocity, to measure progress and determine success. Additionally, VersionOne’s survey data from 2015 reveals that metrics based upon precise quantitative process or business data (such as: cycle time, customer retention, revenue/sales impact and product utilization) are used less frequently on agile projects, despite the fact that they all can be measured. Join us for a discussion to explore the hypothesis that many are questioning the effectiveness of agile methods because too frequently imprecise metrics are used to determine success. Using VersionOne’s survey data, we’ll talk about the importance and potential benefit of shifting to use the metrics that offer greater precision near the bottom of VersionOne’s list to determine success, offer guidance on how to track them, and how to interpret trends observed to make informed decisions. All participants attending will receive a worksheet that highlights the discussed metrics with insights for how to interpret trends observed. The following clip from a July 2015 episode of ThisAgileLife establishes the hypothesis and context for this presentation: https://youtu.be/TmiJQkQMlas