What's the Best Way to Ensure Your Data Analysis is Reliable?
Use a concept compass, keep an eye on your North Star, and you'll go far.
What's the best way to ensure your data analysis is reliable? Point your "concept compass" in the right direction.
This idea comes from Ron Itelman and Juan Cruz Viotti, authors of an upcoming book from O’Reilly, Unifying Business, Data, and Code, who I joined this week on DM Radio.
Here's how a concept compass can guide data analysis:
FIND YOUR NORTH STAR. Interview stakeholders to define the purpose of your journey — the business problem you're trying to solve. Articulate a shared vocabulary, data semantics, and OKRs (objectives and key results).
BUILD A SEMANTIC DATA LAYER. Data analytics is a team sport. A semantic layer unifies business concepts, data, and code and helps your team of travelers get on the same page: business stakeholders, analytics engineers, and data engineers.
DRIVE. REROUTE. RESTRUCTURE. RESUME. Perform your analysis. Adapt your semantic data layer as you go. Reorient your team using your North Star. Keep moving until you reach your destination.
Use a concept compass, keep an eye on your North Star, and you'll go far.
This post was inspired by this week’s DM Radio session with host Eric Kavanagh, and the latest LinkedIn Top Voices in Data Analytics community article: What is the best way to ensure your statistical analysis is reliable? Check it out here.