In the last few weeks, I've been speaking to several folks in our community about best practices in defining clear use-cases as you rollout your data catalog & governance initiatives.
Given how diverse data teams are, you would hear a lot of ideas on how a data workspace like Atlan can be used in the org based on everyones needs and preferences. As a DataOps Champion, in situations like this, it is imperative that you keep an iron-clad focus on ONE specific use-case to begin with.
So in this week's digest I wanted to share with you some best practices and an actionable framework that can help you define clear use-cases & a path to success.
Defining a use-case
Start each use-case with a single user persona in mind. Before you start sketching a solution, take the time to truly understand their pain points. Conduct user interviews/surveys to understand the end-user challenges. For example. you might realise that 80% of the challenges your "data analyst" persona faces could be fixed by simply adding clear column descriptions with guidelines on how to join different tables & columns.
Agile “experimentation” led implementation approach
Embrace an experiment mind-set when implementing a use-case. Accept that each use-case will take a cycle to "maturity" so it's better to start with MVP first with a single team. Remember, it is never too early to involve your end users, so don't forget to include end users when planning & executing an MVP. Use the feedback and learnings from this cycle to roll out the use-case to the entire org.
Ownership and accountability is key
Make sure there is a clear understanding about who are the owners and stakeholders. You need people who have the enthusiasm, band-width, and motivation for driving the project/ use-case (incorporate into OKRs if needed)
Don’t bite off more than you can chew. Prioritisation is key. Pick a problem and solve it end to end.
Data teams are ambitious by nature so it is normal that you would be thinking of multiple amazing use-cases to target. But remember, it is far more valuable to prioritise ONE use-case, totally smash it and then head on to the next one!
With these best practices in mind start putting together a detailed use-case document. The purpose of this document is to ensure everyone is on the same page about the business problems you want to solve, the user personas & success criteria. Eventually, this should become your North Star as you start on your journey of building a DataOps culture.
To help you get started, we have created an easy to modify use-case document template that can help you get started. We even have 2 example use-cases defined.
If you want to read more detailed article is up on our community website: Link
Are you facing challenges zeroing in on your first few use-cases? I'm happy to help over an office hours session. Just shoot me an email and we can set it up!