Uncategorized

<3 Using iPython Notebooks

One challenge I’ve run into as an analyst was in keeping track of analyses, sharing them with my team, and reproducing them at a later time. Despite my best efforts, the analyses I did were often scattered across ad hoc queries and spreadsheets and plotting tools. I would look back at a visualization from months ago and not remember how I got there.

Enter iPython notebooks. When I joined Jana, I was excited to hear that the Jana data team was fully on board with iPython notebooks. The team has a github repository where we store iPython notebooks that various analysts, PMs and engineers across the company are working on. As a n00b at Jana, this was exciting for a couple of reasons:

  1. Onboarding is easier when you can see what kinds of analyses other people have done or are working on. I can see that my co-worker already looked at different ways a product feature could have impacted retention. I can now go back to his notebook, duplicate and build upon it with different techniques or data sources. It’s a great way to learn from your co-workers and internalize product takeaways that others have explored.
  2. It’s easy to rinse and repeat analyses that you know you’re always going to do. With A/B tests, there are a set of business metrics we’ll always to track against. We have a iPython template notebook for experiments where we can just update the experiment name and run it through to get all the plots and tables we’d want in analyzing an experiment.
  3. There’s an active community of iPython notebook users doing cool and interesting things.

You can get started with iPython notebooks for your data team here!

Discussion

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s