Milestone 2: Writeup

Team norms:

To what goals do we all commit?

  • Turn in work on time
  • Come prepared to meetings and class
  • Have fun!

What are our “ground rules”?

  • Assume good intent in all feedback
  • Communicate clearly about availability and time conflicts
  • No super late night meetings
  • Be concise in meetings due to limited time 
  • Be respectful and open to ideas others bring up

What platform will we communicate on? How quickly are responses expected?

  • Communicate by iMessage or in-class
  • Respond by EOD, or sooner if assignment deadline is close

 

AI second pass:

AI categories

  • Customer Requests & Prioritization
  • PM ↔ Engineering Coordination
  • Role Ambiguity / Scope Creep
  • Customer Understanding
  • Time & Attention
  • Startup Uncertainty
  • Tools / Process Friction
  • Technical Constraints 
  • Stakeholder Alignment 
  • Information Overload 
  • Measuring Success / Analytics 
  • Scaling Problems
  • Emotional Load / Stress

Our existing affinity map categories 

  • Communication
  • Responsibility
  • Work Culture / Expectations
  • Streamlining Work
  • Current Strategies

I think the AI caught categories that have more detail and nuance that we didn’t have in the affinity map made by hand. For example, PM and engineering coordination (category generated by AI) and role ambiguity / scope creep (another category generated by AI) I think could both fall under one category generated manually, the responsibility category. The AI was able to break down and find enough quotes and identify two buckets as important independently that we otherwise would have looked over. And as you can see, the AI came up with many more categories than our affinity map had. I don’t think it flattened the transcripts into anything generic. If anything, it seems that it bucketed into more categories than needed and maintained more detail than would be expected for an affinity map. It did not assert anything that wasn’t in the transcripts, though one note about that is that there were certain problems that only came up in one interview, that it created a whole bucket for even though the other interviews didn’t really talk about that.

Avatar

About the author

Leave a Reply