Azure Data Studio (ADS) is a new tool for data professionals from Microsoft. One of the key features of the product are the computation notebooks (or just notebooks). I was tasked with improving the usability and functionality of these notebooks, in order to increase the overall adoption and value proposition of ADS. During the last year, I identified and designed a number of improvements to our notebooks, and many of them were shipped and proved to be successful. Note that due to NDA, I have skipped certain details.
Even though notebooks were considered a core part of Azure Data Studio, the team had received feedback that there were usability issues that needed to be resolved. In addition, the adoption rate of Azure Data Studio was not ideal and needed an accelerator. Due to these reasons, I was tasked with redesigning the notebook experience and thus increasing the value of ADS.
(Computational) notebooks allow people to write code blocks, as well as add text descriptions side by side using natural language. Notebooks are split into cells, which are used to separate various blocks of code and text. The mixing of code and text can be a powerful tool for a number of use cases, including for handing off code to other people, while also including useful descriptions bundled with the code itself, instead of having to create separate documentation files elsewhere.
The key personas that we focused on for notebooks were database administrators and data engineers. These were some of the main target groups of people for Azure Data Studio, and so it was important that our notebooks met their needs effectively.
In order to get familiar with the domain and the state of the art, I conducted an analysis of other computational notebooks. I noticed that there were a lot of different opinions around the user experience of a notebook, and that there were also a lot of potential areas for improvements. My analysis indicated that the field is still relatively new that there are a lot of opportunities for creating value.
In order to get a detailed understanding of the design of the product, I conducted a UX audit. Right off the bat I noticed that the design language used for the notebook did not match the visual language that the team used. Further analysis of the features indicated various potential usability issues that I noted. Though I saw these areas for improvement, I also noticed that the notebooks also had some strengths that I could build upon.
As I dove deeper into the project, I collaborated with our research team to conduct a baseline usability study. Our findings showed that there were usability issues related to core experiences, such as editing content, running code, and the amount of space present.
After collecting all the data, I brainstormed potential solutions that could solve the identified pain points and improve the usability of the notebooks. My lo-fi prototypes were used for gathering feedback and aligning with stakeholders. During the process, we ended up going back and forth and iterated several times as we considered various constraints and options.
As our discussions continued and we started to work through details, we converged towards a more polished design prototype.
We made several improvements to the usability of our text cells. For example, we introdcued a rich text editor, which enabled customers to no longer need to use markdown for formatting. We paired the new editor with a toolbar, which allowed for even greater ease of use. We also introducted a new cell toolbar, which helped with accessibility and discoverability of features.
We also added a new cell toolbar for code cells, which served the same purpose of making these type of cells more accessible and helped highlight important cell-level actions.
We refreshed the notebook toolbar with our design language. This helped make our noteboos visually consistent with the rest of our product. It also introducted a new action order, which made it easier to find the key actions available on a notebook.
In order to help users manage their notebooks easier, we introduced a new Notebooks section to the product. Up to this point, users had to find their notebooks, while going through all the ofther files they had added to ADS. Our new section hid all the unnecessary details and allowed users to focus on their notebooks. In addition, we were also able to introduce notebook-specific search functionality, which allows users to find notebooks or text withing notebooks by keyword.
The redesigned notebook was well received by the team. We were able to ship many of the designs, and we are still continuing to invest in this area in the future. In the following months we heard a lot of positive feedback from our customers. In addition, the overall product adoption accelerated significantly during the last year and a half, resulting in over a 400% increase in monthly active users.