Building an Automated Jira Sprint Report Generation Workflow with Make and LLM: Enhancing Productivity and Optimizing Team Collaboration

Automate your Jira sprint report creation! This guide introduces how to combine Make and LLM to analyze sprint data and automatically generate easy-to-understand reports, dramatically boosting team productivity and optimizing collaboration. This workflow not only saves time but also enables data-driven decision-making.

1. The Challenge / Context

Development teams using Jira spend a lot of time creating reports after a sprint ends. The process of collecting and organizing various information such as sprint goal achievement rate, completed tasks, uncompleted tasks, and issue analysis is repetitive, tedious, and sometimes leads to errors. Especially for teams managing multiple projects simultaneously, the time spent on sprint report creation can be a significant burden. This inefficiency reduces developers' focus on core tasks and acts as a factor hindering overall productivity. Furthermore, inconsistent report formats can make information sharing difficult among team members and impede data-driven decision-making.

2. Deep Dive: Make and LLM

Make (formerly Integromat) is a no-code platform that allows you to build automated workflows by connecting various apps and services. It enables visual configuration of tasks such as extracting, transforming, and transferring data by integrating with various tools like Jira, Slack, and Google Sheets. Make offers a drag-and-drop interface, making it easy for users without coding experience to build automation workflows. Additionally, it provides various built-in functions and operators to facilitate data transformation and processing.

LLM (Large Language Model), especially OpenAI's GPT models, demonstrate excellent performance in generating, translating, and summarizing human-like text by learning from vast amounts of text data. By utilizing LLM for sprint data analysis and report generation, you can create reports that not only list data but also grasp the meaning of the data and provide contextually appropriate explanations. For example, GPT can be used to summarize major issues that occurred in a sprint and suggest improvement plans for issue resolution.

3. Step-by-Step Guide / Implementation

Now, let's look at how to build a workflow that automatically generates Jira sprint reports using Make and LLM, step by step.

Step 1: Create Make Account and Integrate Jira

First, you need to create a Make account and connect your Jira account to Make. After creating an account on the Make website (www.make.com), click "Create a new scenario". Search for and select the Jira module, then click the "Connect" button to link your Jira account to Make. You may need a Jira API token to connect to Jira. Generate an API token in Jira settings and enter it into Make.

Step 2: Extract Sprint Data from Jira

Use the Jira module to extract sprint data. You can search for issues in a specific sprint using the "Search Issues" action of the Jira module. Filter the desired issues using JQL (Jira Query Language). For example, to search only for completed issues in a specific sprint, you can use the following JQL:

project = YOUR_PROJECT_KEY AND sprint = YOUR_SPRINT_ID AND statusCategory = Done

You need to replace YOUR_PROJECT_KEY and YOUR_SPRINT_ID with your actual project key and sprint ID.

The extracted issue data is converted to JSON format for use in Make.

Step 3: Data Transformation and Processing

Before passing the extracted issue data to the LLM, it needs to be transformed and processed into the required format. Use Make's "Iterator" module to iterate through the list of issues and extract necessary information from each issue (e.g., issue summary, status, assignee, resolution time). Use Make's "Text Aggregator" module to combine the extracted information into a single text string. This string will become part of the prompt passed to the LLM.

Step 4: Integrate OpenAI GPT Model and Generate Report

Add the OpenAI GPT module to Make and configure your API key. Use the "Create Completion" action of the GPT module to request text generation from the LLM. In the prompt, specify the desired report format along with the sprint data. For example, you can use a prompt like this:

Analyze the following Jira sprint data and generate a sprint report. The report should include the following:
- Sprint Goal Achievement Rate
- List of Major Completed Tasks
- List of Major Uncompleted Tasks and their Reasons
- Major Issues that Occurred in the Sprint and Proposed Solutions
- Overall Evaluation of Team Performance

Sprint Data:
{{aggregateText}}

{{aggregateText}} represents the text string generated in the previous step. GPT generates the sprint report based on the prompt.

Step 5: Save and Share Report

Save and share the report generated by GPT on your desired platform, such as Google Docs, Notion, or Slack. You can use Make's corresponding modules to automatically save the report and send notifications to team members. For example, you can use the Google Docs module to create the report as a Google document and the Slack module to share the report link in a team channel.

4. Real-world Use Case / Example

Our team develops software using the Agile development methodology. Previously, at the end of each sprint, the team lead had to manually collect and analyze Jira data to write reports. This process took an average of 4-5 hours, adding to the team lead's workload. After building an automated workflow using Make and LLM, we were able to reduce the sprint report creation time to under 10 minutes. The team lead no longer wastes time on data collection and analysis and can now focus on reviewing reports and providing feedback to team members. Furthermore, the reports generated by GPT contained more detailed and analytical content than previous reports, helping team members better understand sprint performance and derive improvement plans.

Specifically, after the sprint ends, a report with the following content is automatically posted to the Slack channel:

• Sprint Goal Achievement Rate: 90% (9 out of 10 stories completed)
• Key Success Factors: Improved code review process, activated pair programming
• Unresolved Issues: Server instability (to be prioritized for resolution in the next sprint)
• Improvement Suggestion: Allocate time for technical debt resolution during sprint planning

5. Pros & Cons / Critical Analysis

• Pros:
  • Time Saving: Reduced sprint report creation time
  • Increased Productivity: Enhanced developer focus on core tasks
  • Data-Driven Decision Making: Provision of detailed and analytical reports
  • Optimized Team Collaboration: Easy information sharing with consistent report format
  • Scalability: Integrates with various platforms and tools
• Cons:
  • LLM Dependency: LLM API call costs incurred, potential for report quality variation depending on LLM performance
  • Prompt Engineering Required: Prompt tuning needed to achieve desired report results
  • Data Security: Security considerations needed when transmitting sensitive data to LLM
  • Initial Setup Complexity: Learning required for building Make workflows and LLM integration

6. FAQ

• Q: Can I build this workflow with Make's free plan?
  A: Yes, you can build a basic workflow with Make's free plan. However, the free plan has limitations on the number of monthly task executions, so you may need to upgrade to a paid plan if your workload is high.
• Q: How do I get a GPT API key?
  A: You can create an account on the OpenAI website (www.openai.com) and obtain an API key. API keys are provided for a fee, and charges are based on usage.
• Q: Where can I generate a Jira API token?
  A: You can generate an API token in Jira settings. Go to your Jira user profile > Account settings > Security > API tokens to generate an API token.
• Q: How can I ensure the accuracy of reports generated using LLM?
  A: Since LLMs generate text based on their training data, 100% accuracy of the report cannot be guaranteed. You must review the generated report and make corrections if necessary. Additionally, it is important to write clear prompts and continuously monitor the LLM's performance.

7. Conclusion

The automated Jira sprint report generation workflow using Make and LLM is a powerful tool for enhancing team productivity and optimizing collaboration. While initial setup requires some effort, automated report generation will contribute to long-term time and cost savings and increase developer job satisfaction. Create a Make account now and build this workflow to revolutionize your team's sprint report creation process!