✉️

#16 Wondering how to prioritize your documentation efforts? Classic 2x2 matrix to the rescue!

Hi, A question I often hear from our DataOps champions is, “How do we decide what to document first?” It is a very valid question. After all, we all have a lot of data assets, have limited bandwidth, and are also usually multi-tasking. So, in this week’s digest, I wanted to share some tips along with a framework to help you prioritize your documentation efforts.

As with any prioritization task, it helps to think of the classic 2x2 matrix:

image

For all the data assets you are evaluating, consider: 1. How easy are they to implement/document? 2. What’s their associated impact? You can use the prioritization template linked here to evaluate the data assets. Your initial focus should be the assets that are quick-wins — i.e they are relatively easy to implement and would also have a high impact. For example, let’s say you have the most popular tables that are used by the data team for customer analysis. This particular set of tables might be the right ones to prioritize:

  • They are often referenced by a team that has a lot of tribal knowledge, which means you won’t have to reach far to find contributors/owners —> High ease of implementation
  • Other teams often reach out to this team regarding questions on customers —> High impact
  • This team is growing and currently, a lot of information resides with existing team members, so having documented data assets will help this team decrease onboarding time for new analysts —> High impact

We’d also recommend thinking about the task of documentation from two perspectives:

⬅️ Looking backward: Documenting existing assets and capturing existing knowledge.

In this scenario, your focus areas may be:

  • Clearing backlog: If your organization has already identified assets with knowledge issues or documentation gaps, then you have a backlog. You can use the 2x2 matrix to prioritize that backlog. You should also make this fun, so leverage tools like gamification. (I know a lot of you have been running very successful gamification drives, and I am constantly amazed by the level of creativity you show 🙌 )
  • High impacts: Another place to start is high-impact assets like BI reports and dashboards. In this case, start by documenting the often challenged figures/KPIs along with the assets involved in their calculations.
  •  The “bus factor”: If you have a single team member who knows everything about an asset, what happens if that person disappears tomorrow? (They could leave your company, for example.) A good rule of thumb is to prioritize assets that have limited or no shared knowledge across more than one team member. These would be “big bet” priorities.

➡️ Looking forward: Identifying “data products” that are in the process of creation, as you want to reduce the burden of looking backward over time.

A detailed article is also up on the community: link

As always, if you have any questions or thoughts, shoot me an email. I'd be happy to brainstorm over an office hours session!

Cheers, Nandini

P.S. The purpose of these emails is to share learnings and best practices to empower our community of DataOps Leaders. All the previous editions of these weekly digests can be found on our community website. Though we have put a lot of thought into curating the most relevant content, if you do not wish to have access to this, you can choose to opt-out by emailing me here: opt-out