The data project pathway: A video newsletter

October 2024 issue

People who work with data are smart: they know how to take a messy dataset and find meaningful patterns that can, potentially, be acted on to help people reach important goals. However, the skills necessary to find insights in data are not the same as the skills needed to make a data project work well. It's too easy to get started on the wrong path, spend time on tasks that aren't productive, and end up with work that doesn't have the impact it should. In my 35 years of working with data, I've made those organizational errors more times than I'd like to admit, and I imagine that you have had that experience, too.

In an effort to help skilled data people avoid these problems, I have proposed "the data project pathway." The pathway is composed of 16 steps (with 2-word names) placed in 4 parts (with 1-word names). I have a webpage that describes it at datalab.cc/pathway. That page includes a list of the steps, a downloadable checklist (as well as a Google Doc), and a YouTube playlist with videos that introduce the pathway and describe each step.

But it's even easier if I embed those videos right here. And, as a result, this will be the first video-centric issue of the #DataIsForDoing newsletter.

And here's to more productive and impactful data work for all of us!

Background

Data project planning models

The Data Project Pathway

Part 1: Planning

Step 01: Define Goals

Step 02: Organize Resources

Step 03: Coordinate People

Step 04: Schedule Project

Part 2: Wrangling

Step 05: Get Data

Step 06: Clean Data

Step 07: Explore Data

Step 08: Refine Data

Part 3: Modeling

Step 09: Create Model

Step 10: Validate Model

Step 11: Evaluate Model

Step 12: Interpret Model

Part 4: Applying

Step 13: Present Model

Step 14: Deploy Model

Step 15: Revisit Model

Step 16: Archive Assets

Conclusion

The Data Project Pathway: Wrapping it up