Tips from product teams to data teams

Recently, I shared views about the decoupling needed in an analytics team to allow self-serve and reserve time to dig tactical and strategic business questions - read more on decoupling dashboards and ad-hoc analyses. Even when teams are decoupled, the issue often remains the same. Analysts' time is a scarce resource! You'll have to pick the right topics to work on.

Before starting Husprey, I worked as a Product Manager. I was expected to carefully prioritize the topics that will drive our roadmap and Business forward. Therefore, I wanted to share two tips to help with your data team's time constraints.

You need specs

Specifications means the topic was considered carefully. It forces people to ask themselves whether they should really ask or do something. It helps set boundaries ("scope" in the jargon) to what is needed and to agree on the success criterion.

When talking about data, requesters should be able to answer those in details :

• Context - Why do we need the asked elements?
• Potential outcomes - What can happen based on the results?
• Recipients - Who will interact with the results?
• Deliverable - How will people interact with the results?
  

More details about those in our article: What makes a good data analytics question.

You need a prioritization framework

Ouch. Don't worry it sounds worst than it actually is.

For the different topics you have in front of you, you need to ask yourself where it stands compare to the others. You need to prioritize. There are different methodologies to do so, Here are two examples: the RICE (Risks, Impact, Confidence and Efforts) one from Intercom is easy to start with! The Gitlab Data team designed one specifically for their needs: the Data Team Value Calculator. The methodology might be easy or complex, what matter the most is consistency!

If you're interested in learning more about how it can be applied to your team, you can reach out so we can discuss it!