Using Data to Measure & Increase Agility in Software Projects
In this inspired talk hosted by Projecting Success, Can Huzmeli goes into detail about the journey of creating the Artificial Intelligence Agile Coach; AIDA, providing insight into how he, and future developers can tackle some of the issues the ICAN team faced.
The talk covers a brief introduction of AIDA, and why it was created, as well as insights into the use of historic project data and lastly the key takeaways of and lessons our team has learned in building an AI product.
AIDA AIDA is integrated with JIRA to track the actions taken within the team, to produce 12 key, measurable metrics which you can learn more about in our alternate blog post "AIDA Assessment Criteria" https://www.icangroup.co.uk/post/aida-assessment-criteria If AIDA spots an outlier in the metrics report and can detect the reason for why it is occurring, AIDA will provide qualitative suggestions on what actions should be taken in response. Why We Did It Digital and Agile Transformation is the basis of ICAN Group, so producing a product that combines these areas was almost a no-brainer.
From working with a number of Agile teams our consultants found many similar comments;
"I have 30+ teams, but our budget can only cover 1 coach"
"What KPI's should I be tracking for Agile teams"
"Agile teams are autonomous but will I know if they are going to fail after months? What if things are going bad and they are not telling me?"
"My teams are working remotely, how can I access their collaboration?"
AIDA in itself is a solution to all of these issues. By using AIDA across your teams you can delegate time effectively while being able to track the progress of each team. The metrics reports AIDA provides are some of the most important KPIs to effectively manage your teams, but there is also space for new metrics at the user's request. AIDA tracks JIRA activities and keeps you informed on your teams' dynamics and project health, the ultimate retrospective tool. How We Use Historic Project Data In our current version of AIDA, we are integrated into JIRA only. Taking sprint and project data we continuously monitor actions (comments, assignees, reporters, the structure of user story) and 42 features are taken into account, allowing AIDA to produce high-quality reports, tailored to your needs. 3 Key Lessons Learned 1. Find Partners not Clients in the pilot launch - the product will need time to adapt, partners will be more understanding of mishaps that occur in the progression to 1.0
2. Supervised learning duration - don't underestimate the workload! (especially where domain expertise is needed) 3. Privacy and Confidentiality - ensure partners are aware and understand encryption