Wait, I've been here before...

1. Lean Startup - Qualitative research for dummies

   One of the most important principles of the lean startup methodology is to validate your assumptions during the problem/solution stage. This is where you have an idea and you need to figure out if someone will actually use/buy it. As part of our new project - SnapCast, we conducted over half a dozen interviews. What do I do with them now? I needed some framework to figure out all this raw data. So I asked a good friend of mine, Yileng Lee, who had studied at the Institute of Design and works as an senior innovation consultant for Doblin/Monitor.

   My email to her was brief:  I need help from you about how to measure and evaluation qualitative data from the interviews. You are a pro at this. Do you have an article or a framework or something practical to follow… I want help to make sense of the interview data.

   Her reply:

    OK, this is a bit of a longer question. Maybe should schedule some time to talk through it. In general it’s messier to analyse qualitative data since you can’t just tally it up and make a chart. 

   High level: 
   
   Collect data 
   you can ask open-ended questions (e.g., “what do you do with your photos today?”,  ”how would you use this tool?” “for the things you said you would use this tool for, what if anything do you use to do those things today?” and/or probe around specific hypotheses “would you want to mash up photos? what things would you want to mash up? why?” ) 
   
   Analyze 
   With qualitative,  you’ll generally get a range of answers and diatribes that are a bit unwieldy to analyze. So you break them down into chunks. That’s what post-it notes are for. Take each chunk of thing people said and put it on a post it note — e.g. “I like to create a collage from an event such as birthday party” That’s analysis 
   
   Synthesize 
   Then look for patterns across all the post-it notes. That’s starting to synthesize. 
   You can do a straight bubble up where you just look for patterns across all the post-it notes. 
   Or you can use a framework to initially sort the post-it notes into first. Generic framework might be AEIOU (activities, environments, interactions, objects, users) 
   In your case you can use your hypotheses as a framework. Sort the post it notes into each of the hypotheses. Which post-it notes relate to which hypothesis? (either supports or opposes it or influences it) 
   
   Interpret 
   Then interpret the patterns. What does it mean that there is this big cluster of comments about x or y. 
   What are the implications for Snapcast — stay high level first instead of going directly to a solution. e.g., Snapcast should make it easier to identify which photos should go together 

   ___________

1 year ago 1 notes
1. currycloud posted this

Theme = Manuscript Neue