Operating Data as a Product Team: Golden Rules — Tip #8
In #7 The Making of a Data Team — 30 Tips to be Data Practitioners, I discussed the formation of a data team. Here, I dive into how I envision a data team’s optimal operation.
Work on Data Products
As we kick off, apart from the outlined roles and responsibilities, it’s crucial to determine our collaboration method. This mirrors the product development process. Instead of rolling out apps or features, we introduce data models and dashboards — which I term “data products”.
Engage in Discovery and Delivery
Although we cater to ad-hoc requests, our goal isn’t to simply produce one-off reports. The Data Analyst (DA) engages with stakeholders to grasp their needs, pains, and desires, guiding the discovery phase. Subsequently, the DA consults on the delivery, i.e., the creation process.
Instead of just providing reports with X columns, we delve deeper with questions with curiosity 🐣 like “Why do you need this data?”, “What decision are you trying to inform?”, and “How does this data assist you?”. The aim is to truly understand the challenge and follow it with informed delivery.
🔖 Benn’s Thoughts on Production
Place Team Members Strategically + Offer Training
Recognizing individual strengths is key:
- Discovery: Determining what to build.
- Delivery: Actual creation.
A DA with an engineering background often has robust systematic thinking and adept SQL skills, making them ideal for delivery. Conversely, DAs with a business inclination understand the perspective of business operators, making them perfect for discovery.
Continuous training remains pivotal. We’ve conducted problem-solving workshops, brainstorming sessions, and weekly retrospectives to enhance our collaborative approach.
Numerous problem-solving frameworks exist, like the 8 Steps to Problem-Solving from McKinsey. However, they all emphasize understanding the problem before leaping to solutions.
Data Product Development Process
It’s paramount that our data customers comprehend that our role isn’t merely about executing orders.
Often, data customers have specific questions or seek insights. While they might have a clear request in mind, a mere data dump might not be the most effective approach. Collaborative problem solving between the data team and customers will yield more meaningful results.
🤜 Discover how others achieve this
Too Idealistic? 😎
I acknowledge that this approach might sound utopian. Perhaps it’s my product manager background influencing this perspective. This method aligns with how I collaborate with product teams, and it’s proven effective. In upcoming articles about the Team and Company, I’ll share my experiences trying to implement this approach, discussing both triumphs 👍 and setbacks 👎.
See others who advocate the same mindset
🤩 I’m happy to hear how you handle product development process? what tool do you use? Feel free to reach out to me on LinkedIn Karen Hsieh or Twitter @ijac_wei.
🙋🙋♀️ Welcome to Ask Me Anything.
