Background: Previous research highlights that common misconceptions about developer productivity lead to harmful and inaccurate evaluations of software work, pointing to the need for organizations to differentiate between measures of production, productivity, and performance as an important step that helps to suggest improvements to how we measure the success of engineering teams. Methodology: Using a card sort, we explored how a Three Layer Productivity Framework was used by 16 software engineers at a Software Engineering focused conference to rank measures of success, first in the current practice of their organization and second in their individual beliefs about the best ways to measure engineering success. Results and discussion: Overall, participants preferred organizations to 1) continue their prioritized focus on performance layer metrics, 2) increase the focus on productivity metrics, and 3) decrease their focus on production metrics. When asked about the current metrics of their organizations, while all roles reported a current focus on performance metrics, only ICs reported a strong focus on production metrics. When asked about metrics they would prefer, all roles preferred more performance metrics but only leaders and ICs also wanted productivity metrics. While all participants were aligned on performance metrics being a top preference, there was misalignment on which specific metrics are used. Our findings show that when measuring developer success, organizations should continue measurement using performance metrics, consider an increased focus on productivity metrics, and consider a decreased focus on production metrics.

翻译：背景：先前研究指出，关于开发者生产力的常见误解会导致对软件工作的有害且不准确的评估，这意味着组织需要区分生产、生产力与绩效的衡量指标，这是推动改进工程团队成功评估方式的重要步骤。方法：通过卡片分类法，我们探究了16名软件工程师如何在软件工程专题会议上运用三层生产力框架，分别从组织当前实践与个体对最佳工程成功衡量方式的认知两个维度，对成功指标进行排序。结果与讨论：总体而言，参与者更倾向于组织应：1）继续优先关注绩效层指标，2）增加对生产力指标的关注，3）减少对生产指标的关注。在询问组织当前使用的指标时，尽管所有角色均报告当前侧重于绩效指标，但仅有个体贡献者报告对生产指标的高度关注。在询问其偏好指标时，所有角色均偏好更多绩效指标，但仅领导者与个体贡献者同样希望增加生产力指标。尽管所有参与者在绩效指标为首要偏好上达成共识，但在具体使用哪些指标上存在分歧。研究结果表明，在衡量开发者成功时，组织应继续使用绩效指标，考虑增加对生产力指标的关注，并减少对生产指标的依赖。