ZoomInfo 数据团队如何用 DPRD 与五大角色构建现代数据团队
Data Product Teams : Best Practices for a Modern Data Team
ZoomInfo 数据副总裁 Philip Zelitchenko 分享了其团队的数据产品实践：用 DPRD（数据产品需求文档）定义数据产品，并配套 TEP 技术执行计划对齐架构与治理。团队设数据 PM、业务逻辑、数据分析、数据治理、数据平台五类角色，并正在构建内部产品 Heartbeat，自动衡量主要数据产品的使用情况并评估优先级与 SLA 影响。
In short : Explore best practices for modern data teams, focusing on roles, data products, and efficient scaling strategies.
Data teams are becoming software engineering teams.
On December 14th we welcomed Philip Zelitchenko, VP of Data from ZoomInfo, to talk about how he has built this discipline within his team & it was fascinating. The video is here.
Like the Devops movement, the Dataops movement aims to scale the use of data within companies without increasing the headcount of the data team.
To do that, Philip defines data products using DPRDs, structures his data team with five key roles, & defines clear roles between the data team & others in the company.
DPRDs, or Data Product Requirements Documents, contain the key information about a data product: what it will provide, how it will produce value, how the data will be governed including data quality alerting.
Unlike code, data is stochastic or unpredictable. Data may change in size, shape, distribution, or format. This adds an additional dimension of complexity to the DPRDs.
In addition to the DPRD, the ZoomInfo data team employs TEP or technical execution plan that aligns the internal technical teams on architecture & governance.
The data team has five key roles:
- Data PMs : quarterback the DPRDs. They gather feedback from users, define the value, solicit feedback from the rest of the team, then manage the execution of the plan.
- Business logic : the data engineering team build the ETL pipelines while the data science team researches & implements machine learning algorithms for ML\DS driven data products.
- Data analysts : embedded/seconded to the different operating teams, analysts analyze the data each team needs using the infrastructure provided by the data platform.
- Data governance : ensures data quality/accuracy, defines the access control policies for security, sets the operating procedure for alerting & monitoring, and help define data contracts between producers, processors, and consumers.
- Data platform : builds the universal data infrastructure for the company.
Last, the ZoomInfo team is building an internal product called Heartbeat that measures usage across the main data products, evaluate the priority, SOPs for impact on SLAs and communication with data practinioers across the org in an automated way.
For Philip, leading the data team is about focusing on the data products that drive meaningful value to the company. I learned a tremendous amount about the way modern data teams, who leverage software engineering disciplines, operate.
Thank you, Philip!
Get the next one in your inbox
The 1-minute read that turns tech data into strategic advantage.
Read by 150k+ founders & operators.
GP at Theory Ventures. Former Google PM. Sharing data-driven insights on AI, web3, & venture capital.
来源：Tomer Tunguz 博客 · tomtunguz.com