What’s a data notebook and what is it used for?

Ever since the release of project Jupyter in 2014, data notebooks have gradually become a standard tool for data exploration. Although data notebooks have mainly been used by technical teams so far, new tools have recently appeared. They are well suited for data analysts’ hard skills while being accessible for Business Users, enabling them to run through sharp insights. Here is why and how data (and maybe business) teams should use this revolutionary – so to speak – tool.

‍

What is a data notebook?

Definition

Data notebooks are a form of interactive computing in which users can write and execute code, visualize the results, and share insights with their team.

Data notebooks offer a wide range of benefits:

• they enable you to explore your databases via several Big Data processing technologies;
• they give you immediate use of your data thanks to built-in visualization modules and various output mechanisms;
• they ensure full interactivity, allowing editors and viewers to read analyses more effectively.

Data notebooks vs. dashboards

Dashboards help Business Users get a better understanding of their day-to-day operations, by enabling them to:

• monitor metrics;
• get an overview of several KPIs;
• notice dynamics;
• looking at the data from a specific perspective (an event, a given period of time).

But dashboards also have significant disadvantages, such as:

• high maintenance costs in order to fix broken queries and widgets, add new filters or improve the UI;
• impossibility of evaluating past choices, tracking informed decisions, or building up related knowledge;
• lack of context, which prevents Business Users from knowing what a particular dashboard has been created for, and thus limits their understanding of the trends and figures identified.

These two tools shouldn’t be considered mutually exclusive. Dashboards are great for monitoring, while data notebooks are a must-have for data exploration and reporting: make your choice in accordance with your business goals!

What are data notebooks used for?

Data scientists have been using data notebooks for quite a while. They often use the Jupyter notebook, along with Python coding to run data exploration tasks. But a new generation of notebooks – such as Husprey’s – have been building improved ways to empower Data Analysts.

Here is how a Data Analyst can make use of a notebook:

• Ad-hoc analysis: Analysts can use notebooks to provide Business Users with specific information (concerning churn, margins, costs, etc.) and share key insights with non-technical teams.
• Collaboration and documentation within the data team: notebooks enable Analysts to leverage the knowledge that the team has collected and improve global productivity in the long run.
• Onboarding: notebooks give new Analysts easy access to former analyses, which allows them to quickly acquire the skills and knowledge they need.

Notebooks provide analysts with a strong technical environment, which gathers context information, data intelligence and SQL queries into a single tool. Iterations among members of data and business teams are much faster thanks to their user-friendly interface.

Throwback

Before notebooks, Analysts used BI tools and dashboards. Using screenshots, they copy/pasted their tables and charts into Notion or a Google Docs file. This whole process was very time-consuming, and damaged the data’s quality and traceability.

A crucial tool for successful Data Storytelling

Making the most of your data means bringing storytelling into the mix. Data notebooks combine your data and its context with some reasoning, interpretations and queries: this makes them the perfect fit for any project that requires storytelling, whether it be investigation, exploration or post-mortem evaluation.

A notebook-based collaborative tool enables you to give your data and comments a story format, in order to display your logical train of thought and arguments. You also have the possibility to highlight your assumptions and their limits, as well as to present and explain your interpretation of the results.

Keep in mind that your insights must be clear, concise and actionable in order to actually help you drive business outcomes.

Here are 3 key features that will make your story striking and powerful:

• An effective summary: to give Business Users an overview of your insights from the very beginning, start by introducing your key findings and suggestions, along with some context and a sufficient level of detail.
• Written documents: they give a strong structure to the analysis, more effective than slides. Data Analysts focus on the content rather than the format.‍
• Visualization: creating and designing charts will make your arguments more striking and noteworthy. Don’t forget to add legends and context information: they can substantially improve your document’s clarity!

A data notebook enables Analysts to explore databases, write queries and create stories, with a single tool. But data notebooks are also suited for Business Users. The UX interface of it empowers them to understand, comment and take decisions from data. It's a real game changer on both technical and collaborative sides.