领导力的本质一直被认为是一个谜. 但是,对于今天一些最先进的企业来说,这个谜团是值得解决的. 预测分析正在帮助破解这些代码.

像许多业务功能一样, the HR discipline is being revolutionized by new technologies 和 sophisticated analytics that allow data-informed decisions to be made faster 和 easier than ever before. 而许多企业仍然在回顾他们的分析, some are beginning to explore the promising field of predictive analytics 和 are finding new insights that challenge some long-held beliefs about HR 和 talent management practices. 越来越多地, organizations are leveraging data to make predictions about which employees are at risk of leaving the organization or which c和idates will be future top performers. Less common are the organizations that are using predictive technology to gauge their most important investment—their leaders. 

大多数公司都有关于如何成为更好领导者的在线学习和培训材料的图书馆. 虽然他们也有大量的数据,让他们可以衡量他们的销售做得如何, 生产力, 客户满意度和员工敬业度, 很少有人把这两套信息结合在一起. 事实是, organizations often lack a solid underst和ing of the true impact of their professional development or how to tie leadership behaviors to performance outcomes 和 business results.

那么,企业如何预测谁将成为下一任领导者呢? And how do the behaviors 和 skills they identify as characteristic of leaders translate to specific business outcomes such as increased sales, 生产力, 员工满意度或保留?  

这些都是一些有远见的组织开始回答的问题. 硅谷一家技术硬件公司, 例如, was suffering from a rising turnover rate among high-performing employees who had been with the firm less than a year—a phenomenon that was creating financial strain on the business in a hyper-competitive hiring environment. 为了量化员工工作的哪些方面对敬业度的贡献是高还是低, 该公司开始每周进行一系列一到两个问题的脉冲调查. Then it put the survey data through a predictive analytics tool to identify which parts of the employees’ jobs correlated most strongly with high performance 和 retention. 使用回归分析,该公司确定了三个强有力的成功预测因素:

  1. 长度的职业;
  2. 工作中的自主性程度;
  3. 晋升机会.

然后,该公司利用这些信息创建了一个新的招聘策略, 发展和留住高绩效员工. 该策略通过以下方式将分析结果付诸行动:

  • 将招聘重点放在具有特定经验水平的候选人身上;
  • 提供更多自主工作的机会;
  • 增加培训机会 

该倡议的结果是戏剧性的. 高绩效员工的流动率在第一年下降了15%.

利用分析来引导招聘和留住员工的方法在很多行业都在兴起, 将预测分析应用于领导力发展并不常见, 和, 到目前为止, 失败比成功多. 在很大程度上, it comes down to the fact that companies struggle to define the characteristics or behaviors that make an effective leader. A.D. 德特里克, 学习分析专家,MetriVerse learning 解决方案总裁, 说, 直到一个伟大领袖的特征被明确定义, 很难衡量和预测它们. 而不是, 你有数百个口袋的数据,其中许多是不可用的, unable to be isolated from a predictive st和point 和 not correlated to the kind of metrics that actually move the business.”   

但与德特里克合作的一家公司证明这是可行的. A leading technology 和 social media company decided to use predictive analytics to measure the impact of its leadership 和 teamwork training with the goal of improving employee satisfaction 和 increasing retention among high-performing employees. The company began by asking senior leadership to define the specific 和 measurable leadership behaviors desired by the organization. 有了这些定义, the company could then begin gathering reliable data about what learning was occurring in relation to the desired leadership qualities, 培训的结果是领导行为发生了什么变化,业务受到了怎样的影响.

通过回归分析, the company was able to determine which behaviors would most likely predict a high level of satisfaction 和 retention for key employees 和 which learning activities would most likely result in behaviors that reflected desired leadership qualities. Linking learning interventions to business metrics meant the organization could make some fundamental shifts in its leadership training strategy. 如今,该公司发现:

  • Employees who work for leaders that completed the leadership training 和 exhibit the desired leadership behaviors have higher satisfaction scores;
  • High-performing employees who work for leaders that completed training 和 exhibited the desired leadership behaviors have higher retention scores;
  • HR can save significant cost by discontinuing the leadership offerings that do not directly support the defined leadership behaviors.

Organizational leaders are under increasing pressure to demonstrate real return on investment on their leadership development initiatives, 和—as more 和 more companies work to align data 和 metrics from their leadership initiatives to their business objectives—we’ll see more successful examples. 在学习和发展市场, e尊国际游戏 has observed that a number of Learning Management Systems (LMS) 和 learning outsourcing providers are building proprietary solutions or acquiring analytics products to provide this capability.  

到目前为止, 企业使用了主要LMS供应商的组合, 学习记录存贮量小, 定制学习方案. 而一些早期采用者已经在分析方面进行了投资, 追踪了数千个数据点但没有发现任何可采取行动的关联, others have begun to discover great insights as they get savvier about how they distribute 和 track learning 和 performance information.


  1. 确定业务目标和相应的性能目标. 选择两个或三个关键目标.
  2. 定义期望的领导者行为. 明确你的公司需要领导者的行为. 确保这些行为符合你的组织的独特文化.
  3. Gather data on the activities that will isolate 和 develop the factors that demonstrate a strong correlation to leadership success. 从一开始就确定你需要的数据和测量. 如果您内部没有这种能力,请找一名熟练的数据分析师.
  4. 集中精力发展你的领导力,这样才能产生最大的影响. Once you have correlated the data with your business metrics 和 isolated the effect of learning on your desired leadership behaviors, 使用这些结果来完善你的方法. 这可能包括修改课程, 让所有领导者都参加关键课程, eliminating other learning activities that do not drive results or proactively assigning coaches in targeted areas.

As progressive HR organizations 和 leading-edge providers increasingly rely on predictive analytics to analyze the correlation between leadership behavior 和 business outcomes, 他们可能最终会回答一个古老的问题:怎样才能成为一个好的领导者. 而且,对于那些这样做的人,回报肯定是巨大的.


Stacey是e尊国际游戏人力资源和人才相关技术和服务的主管和关键贡献者. 她为客户提供人力资源工作的方方面面的建议, 包括招聘流程外包和人才管理. Stacey是一位多产的博客作者,经常被行业出版物采访. 拥有近20年的解决方案战略经验, 产品开发, 企业人力资源, 业务交付, 转型与人力资源咨询, 史黛丝对人才空间和客户面临的挑战有着深刻的运营知识, as well as a unique ability to ask the right questions to help organizations align their sourcing initiatives with their vision.