One team doing client research this week asked if 30 interviews would be enough. We said 30 seemed like a reasonable number as long as it included sufficient concentration and diversity. That is, if each one of the 30 people had a different demographic, a different reason for using or not using the product, a different attitude – liking or hating – the product in its current form, and different alternatives to using it, then there might not be a large enough concentration of any particular findings to draw your attention. You need to find some patterns in your research in order to help you generate and choose a solution. But you also want diversity to help you define your target client and understand whose problems you won’t be solving and why. So, picking interview subjects may need to be an iterative process – seeing what you get before choosing more.

Julie Norvaisas, of Portigal Consulting, posted a presentation that comprehensively addresses how to pick the right interview subjects. Her firm typically works within similar time constraints to our innovation leadership programs – about three-and-a-half weeks from first interview to opportunity.

Here is the link to her full presentation. I’ve extracted some of the practical advice and added some in [ ]:

• Get over your hesitance to ask people. “Most people are curious and eager to be part of something that contributes to progress” [As long as you clearly distinguish the research you are doing from sales]. People rarely get asked to talk about any aspect of their lives in detail – it’s flattering.
• Establish criteria
  • Relationship to the product/service: Let your business questions guide your selection of “typical user, non-user, extreme user, peripheral user, expert user, subject matter expert, wanna-be user, should-be user, future user, hater.” Select some for contrast, e.g., haters and extreme users.
  • Type: Consider the whole system “chooser, influencer, user” and the relationships between them. Think about who you need to design for.
  • Context: Where would they use it? What would they use it with/before/after?
  • Demographics: Where could you possibly test your solution first? Who could possibly afford to use it
• Screen
  • Have initial questions (prior to a full interview) that identify subjects who “have a point of view, care about the thing, and are articulate.” Don’t waste time with people who can’t or don’t want to help you
  • [People are more inclined to help if they get something back such as a summary of your findings or a thank you gift or public acknowledgement of their contribution]
  • Make them feel special and make the interview something they want to do. “I want to invite you…”, “I need your help…”, “You can create something better…”, “You are one of a select few…”
  • Write out a standardized screener for all interviewers to use so you can check off the criteria
• Get help
  • [reach out to others who could share the work or trade interview subjects or share findings]
  • [leverage research already done]
  • [ask others to make introductions]
  • Ask the person to recommend others (snowball recruiting)
• Build rapport prior to the interview
  • [Participants in our programs let people know – that it’s part of a learning experience]
  • Ask for special conditions you need to be aware of
  • Confirm: how long it will take, how they should prepare, what they will be asked, who will see their responses
  • Thank them profusely