提问是组织内用于激发思考和收集信息的关键工具。凯伦·黄的研究表明，提出更多问题的人通常更受欢迎，提问也是衡量职业晋升的一项指标，在董事会中成为领导者的主要产出。缺乏提问往往预示着一种不接受意见的文化，当求职者认为面试过程没有挑战性时，他们更有可能拒绝高薪的工作邀约。经受严厉的提问是一个双向筛选工具；温和的问题会被求职者解读为关于未来同事质量的负面信号。

问题的有效性在很大程度上取决于情境，但一些通用规则同样适用。为了建立人际关系，提问者必须真诚地倾听回答，而不是进行“回力镖式提问”（即仅仅为了分享自己的信息而提问）。研究人员艾莉森·伍德·布鲁克斯和迈克尔·约曼斯发现，回力镖式提问被认为是不真诚的。此外，在进行面试求职者或资助创始人等比较评估时，一致性至关重要。阿米莎·米勒的研究表明，通过标准化评估框架，可以消除甚至逆转投资者提问中的性别偏见（即男性被问及增长，而女性被问及风险）。

然而，向每个人都问相同的糟糕问题并没有用。研究人员南迪尔·巴蒂亚、蔡伟和萨米尔·斯里瓦斯塔瓦利用大型语言模型来识别“变化球”问题，他们将其定义为难以预测和准备，但高度相关且难以回避的问题。通过分析美国上市公司的电话会议记录，作者发现分析师提出的变化球问题导致了更大的股价波动和评级变化。研究还显示，明星分析师和新开始覆盖某家公司的分析师最有可能提出变化球问题，因为前者依托深厚的专业知识，而后者未受传统观念的束缚。

Questions are crucial tools within organizations for sharpening thinking and gathering information. Research by Karen Huang suggests that people who ask more questions are generally better liked, and questions also serve as a measure of career progression, becoming a leader's principal output in the boardroom. A lack of questions often signals an unreceptive culture, and candidates are more likely to turn down high-wage job offers when they rate the interview process as undemanding. Undergoing tough questions serves as a mutual screening tool; softball questions are interpreted by candidates as a negative sign regarding the quality of future colleagues.

The effectiveness of a question depends heavily on the context, but some general rules apply. To build relationships, questioners must genuinely listen to the answers rather than "boomerasking"—asking a question simply to share one's own information. Researchers Alison Wood Brooks and Michael Yeomans found that boomerasking is perceived as insincere. Additionally, when making comparative evaluations like interviewing job candidates or funding founders, consistency is vital. Research by Amisha Miller showed that gender bias in investor questioning, where men are asked about growth and women about risks, could be eliminated and even reversed by standardizing evaluation frameworks.

However, asking the same poor question is not helpful. Researchers Nandil Bhatia, Wei Cai, and Sameer Srivastava used large language models to identify "curveball" questions, which they defined as difficult to predict and prepare for, yet highly relevant and hard to dodge. By analyzing earnings-call transcripts of listed American firms, the authors found that curveball questions from analysts led to larger share-price movements and rating changes. The research also revealed that star analysts and analysts new to covering a firm were the most likely to ask curveballs, as the former draw on deep expertise and the latter are unburdened by conventional wisdom.

Source: The secret to good questions

Subtitle: Consideration, consistency and curveballs

Dateline: Jul 09, 2026 05:22 AM