Enhancing Jira's Health with Site Optimizer

Have you ever thought about why some products prompt you to free up space or why they have a storage limit anyway? Read through for the answer.

But first, shoutout to my colleagues:

Impossible alone, possible together

???? I am working from the India office alongside another Senior PM. Over the past 4 months, a team of 7 engineers, 1 designer, 1 data scientist, and 1 marketer collaborated from various locations such as Australia and the USA to build Site Optimizer.

Let’s dive right into the journey. ?

Part 1: The Data lifecycle

Data Lifecycle refers to the phases data goes through from initial creation to sunset.

Any data moves through 3 phases in its lifecycle:

1. Initiate aka The creation phase: In this phase, the user creates new data for eg. Creating new issues/tickets in Jira (or Clicking a new photo from your phone)
2. Active phase: In this phase, the users are actively using and interacting with this data for eg. Viewing or editing the issues. (similar to sharing or viewing the photo)
3. Sunset or Clean up phase: The end of the data lifecycle is sunset (soft delete or permanently delete). In this phase, the data is not required anymore and should be removed. If not removed, data volume endlessly increases before the user starts witnessing poor performance. For example: In Jira, if customers go beyond 5 million issues, they’d experience performance and productivity impact. (similar to going back 5 years to delete the old photos)

Part 2: The Customer Research

As a PM, one thing we’re always striving for is to enhance product performance for our customers. While building scalable systems guarantees better performance, our team at Jira Platform learned that the customers are also interested in having tools to manage the speed and health of their site.        

It was a huge learning for me when I got to conduct/participate in over 20 interviews (sometimes even independently!) with some of the largest Jira customers. The key insights that we uncovered:

2.1 - Customers want to clean up data

1. Our customers are seeking tools to independently manage the health and performance of their systems.
2. Many of these customers want to clean up their data in the upcoming 12 months.
3. Admins (one of the user personas) are responsible for cleanups as part of their administrative duties to ensure a healthy Jira site for everyone.
4. Admins conduct clean-ups at least 1-2 times annually.
5. Additionally, some customers developed in-house solutions to accomplish this objective.

2.2 - But there are some challenges...

The 2 major problems that showed up:

1. Lack of insights: Jira Admins didn’t have clear information or insights about the data and its usage. They are unclear about What

1. to clean up and How much to clean up.
2. Mitigation actions: Jira Admins lacked user-friendly mitigation strategies, such as bulk clean-up with a single button. This led to an increase in the time and effort spent on manually searching data across Jira and then cleaning it up one by one. Similar to deleting a photo

These insights underscored the significance of a tool/solution.

2.3 - Also, the data indicated,

Further, The data analysis indicated that customers experiencing poor performance often have large amounts of data. A significant portion of this data is unused i.e. either not been looked at or not edited in 2 years. Hence, can be easily moved to the third phase i.e. the clean-up phase.

Hence, our goal was to improve the performance of Jira for our customers by empowering them to clean up old and unnecessary data.        

?? Part 3: Welcome Site Optimizer

Site Optimizer is a hub for Jira admins, providing valuable insights into their site's health. It enables them to do bulk cleanups, including issue archival and custom field deletion, which will improve their site’s performance and hence their productivity.

It is an enterprise edition feature.        

1. Insights i.e. What data to clean up and by how much.
2. Actions i.e. Ability to bulk clean up the data
3. Recover i.e. The ability to restore the data.

Insights

In this release, we’ve shipped 3 attributes - Custom fields, Issues and Project role permissions. For each attribute, Site Optimizer provides the following insights:

1. Total Data volume for an attribute on your site eg. 3220 custom fields in the example below.
2. Recommended maximum i.e. the suggested volume beyond which Jira site might experience performance regressions. For eg. 6000 is the recommended limit for custom fields.
3. % usage i.e. (Total data volume)/(Recommended maximum)*100 to showcase how close the Site is to the Recommended maximum, in this case 6000.
4. Last action taken i.e. the date of the last clean-up action taken.

Clean up Actions

For clean-up actions, the Site Optimizer provides soft delete i.e. Archival or Trashing. The admins will have 60 days until the custom fields are permanently deleted.

Soft delete is important for customers with data retention policies. It allows them to clean up their site while having an option to manually restore if required.

Recover

The data cleaned up by the admins is available in the activity log for 60 days (for custom fields) from where it can be recovered. The activity log also provides an option to download the removed data as a CSV.

While this is just the beginning, the Journey of Site Optimizer stands as a testament to what can be achieved with a dedicated team and a clear vision. Within, 4 months the team moved quickly and brought a step change in the lives of admins, our largest customers and their Jira experience. ???