For over a year, software giant Atlassian has been investing heavily in AI- and machine learning-powered tools across its portfolio. Today, the company detailed the recent fruits of its labor, revealing that it has analyzed patterns from over 174,000 customers to understand “the bigger picture” behind team interactions.

“The Atlassian platform is making over 17 million predictions every day through smarts and machine learning,” head of product Shahib Hamid said. “Using these insights, we can provide … recommendations within products customers use throughout their whole workday. More recently, we began developing predictive, smart experiences in our products to help make teams more productive.”

Atlassian Smart Search

Above: Atlassian’s smart search feature.

Image Credit: Atlassian

Confluence and Jira recently gained improved document search via an AI component called smart search. Atlassian claims it raises the chances of finding a file relevant to a query by 33% while complementing instant search results, a module that surfaces predicted search results before users type a character. Dovetailing with smart search and instant search results are intelligent filter controls, which anticipate filters a user is most likely to choose in order to narrow down a search’s scope. According to Atlassian, when these filters are implemented, users select them 89% of the time.

Elsewhere, a new web dashboard called Start shows a personalized overview of any Confluence and Atlassian projects a customer has worked with before. In Jira and Confluence, predictive user mentions recommend a list of people to bring into a project. And predictive user pickers takes this idea one step further by suggesting relevant teammates to collaborate with in various scenarios across Atlassian products.

Issue assignment in Jira is now more predictive, as well, with the ability to know who’s active on a project and who works on similar issues. (Atlassian says its algorithms can project the five most likely assignees with 86% accuracy.) In Confluence, AI-powered page restrictions can identify who users collaborate with and what they typically work on to recommend who should be restricted from viewing a page. And Bitbucket can now predict the best reviewers for a pull request based on similar pull requests in the past.

Atlassian Cluster Similar Tickets

Above: Clustering similar tickets in Jira.

Image Credit: Atlassian

Beyond all this, Jira will soon get a new feature that allows users to cluster similar support tickets. According to Atlassian, this predictive technology is already used in Jira Software to group similar bug reports or features requests, linking incidents to tickets in Jira Service Desk and displaying related knowledgebase articles in Confluence. (Atlassian claims it predicts certain fields within Jira, including versions, labels, and components, with between 75% and 79% accuracy.)

“With hundreds of service desk tickets to get through each day, triaging similar tickets all at once can be a huge time saver for IT teams,” Hamid said. “By learning from historical data, we can make many fields in Jira Software intelligent. When filling in certain components, labels, and versions of a product, predictive fields immediately surface the most relevant ones.”

How startups are scaling communication: The pandemic is making startups take a close look at ramping up their communication solutions. Learn how