Scholarlink

<p>We introduce the Gryzzly time-tracking dataset: a longitudinal, high-resolution collection of 4.4 million interactions recorded between 12,447 users and 173,323 tasks across 50,759 projects, spanning from 2017 to 2024. Compiled from real-world usage data of the Gryzzly software, the dataset encompasses projects from diverse industries such as marketing, finance, and banking. It provides a detailed view of daily activities contributing to project completion, including information about the users involved, the tasks they worked on, and the planned versus actual costs of each project. To validate the published data, we analyzed the underlying temporal collaboration network, revealing expected patterns such as circadian user activity, power-law characteristics in degree distributions, and heterogeneously distributed inter-declaration times. Additionally, we observed well-documented failure dynamics, including a heavy-tailed distribution of failure streak lengths and diverging performance improvement trends between successful and failed projects. These features make the Gryzzly dataset a key resource for studying productivity, team dynamics, and project failure.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Seven years of time-tracking data capturing collaboration and failure dynamics: the Gryzzly dataset

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年4月5日

查看全文
  • Jacob Levy Abitbol,
  • Louis Arod

摘要

We introduce the Gryzzly time-tracking dataset: a longitudinal, high-resolution collection of 4.4 million interactions recorded between 12,447 users and 173,323 tasks across 50,759 projects, spanning from 2017 to 2024. Compiled from real-world usage data of the Gryzzly software, the dataset encompasses projects from diverse industries such as marketing, finance, and banking. It provides a detailed view of daily activities contributing to project completion, including information about the users involved, the tasks they worked on, and the planned versus actual costs of each project. To validate the published data, we analyzed the underlying temporal collaboration network, revealing expected patterns such as circadian user activity, power-law characteristics in degree distributions, and heterogeneously distributed inter-declaration times. Additionally, we observed well-documented failure dynamics, including a heavy-tailed distribution of failure streak lengths and diverging performance improvement trends between successful and failed projects. These features make the Gryzzly dataset a key resource for studying productivity, team dynamics, and project failure.