AI Tagging in Jira and Zendesk

[Melissa Carelli]

Hello Atlassian Community,

I’m interested in automating the way our team determines and assigns root causes to support tickets in Zendesk and in Jira. Specifically:

1. Zendesk: Is there any built-in functionality or recommended integrations that leverage AI or machine learning to analyze customer interactions (e.g., ticket descriptions, conversation text) and automatically apply a root cause or category?

2. Jira: Likewise, are there any AI-driven plugins or features within Jira (or integrations with Jira) that can analyze ticket content and automatically fill in a “root cause” field or similar classification?

Our goal is to reduce manual effort and ensure consistency in how we categorize and address issues. If you’ve had success with a specific app, integration, or workflow, I’d love to hear about it. Thanks in advance for any guidance or recommendations!

[Devin Booker]

Hello,

Automating the process of assigning root causes to support tickets is a great way to enhance efficiency and consistency in your workflow. Here are some suggestions for both Zendesk and Jira that you might find helpful:

Zendesk:
Zendesk has several built-in features and integrations that can assist with categorizing tickets. One option to consider is Zendesk's Answer Bot, which utilizes AI to analyze ticket conversations and suggest appropriate articles from your knowledge base. Although it doesn’t directly assign root causes, it can inform your team about recurring issues that could be categorized.

For deeper analysis, you might explore third-party integrations like MonkeyLearn or Zendesk’s own Explore analytics tool which can potentially be trained to identify patterns in customer interactions. These tools can help in tagging tickets with relevant issues based on historical data.

Jira:
In Jira, there are plugins such as Automation for Jira which can introduce rules that help classify tickets based on keywords or phrases found in the description. Moreover, look into Smart Issue Types or Jira AI, which leverage machine learning to suggest issue types and potentially root causes based on previous tickets and their resolutions.

Additionally, tools like Scriptrunner can help you create custom scripts that analyze ticket content and automatically fill in fields based on defined criteria.

For your specific use case, consider a combination of these tools to create a streamlined workflow that minimizes manual effort.

If you’re looking for a business that specializes in custom integrations and automations, we would love to help you optimize your processes. Our team can work with you to implement tailored solutions that ensure seamless integration of AI capabilities in your support systems.

[Payal _appbox_ai_]

Hi Melissa, 

Welcome to the Atlassian community!

We’ve developed a native Jira app, GPT for Jira, that leverages AI to categorize issues based on their descriptions. You can bring your own OpenAI API key and customize AI prompts to fit your needs, ensuring the AI works within the context of your tickets.

For example, in the scenario below, issue categories are documented on a website (which could also be a Confluence page). The GPT application analyzes the issue description and determines (using your confluence page) that the user’s problem falls into the Mobile Compatibility Issue category.

https://marketplace.atlassian.com/apps/1236030/gpt-for-jira-pro?tab=overview&hosting=cloud

The prompt used in this example is:

I will give you the issue the customer is experiencing, using the following website, you have to categorize the issue in 10 words or less. Just give me the category, nothing else. Website - https://websites911.com/blog/common-website-errors/ Issue:"${issue.description}"

We also have a feature pending to "persist" the AI response in a custom Jira field.

If you are interested in this, we would be happy to connect and chat further. 

Thanks,

Team Appbox.