The trick to creating a good conversation is having a good conversation guide and then being prepared to be authentic and allow the conversation to wander wherever it goes. One way to assess the authenticity of a conversation would be in terms of the number of times you (and your conversationalist) are surprised by it, or find new ideas in it.
In an authentic conversation the participants are equally able to lead (or block) any avenue of exploration. Your conversation guide must then provide you with a good idea of the avenues you want to go down and why.
To get there you need to move away from your product / commercial / technical focus and reframe your goals in terms that make sense to your user or customer.
Some Principles
- Think like a detective. Solicit spontaneous, unprompted comments first. Plan out how you are going to gradually move from open-ended to explicit, direct questions.
- Focus on specifics. “What did you do yesterday?” will get better conversation than “What do you do on a typical day?”
- Take the person to the relevant context (at the very least in their imagination) – “tell me about the best gaming experience you’ve ever had?”
- Like an identity parade, ensure you have multiple alternatives for them to consider – not just your product concept of interest.
The steps to getting there
- List out the decisions that you need to be able to make from a product / business / technical perspective.
- For each of these decisions consider what questions you would need to know the answer to in order to make the right decision.
- These questions are almost certainly not meaningful to your users / customers and so the next step is to work out what there is in their perspective that might help you work out the answers to your questions. In effect, you need to treat the conversation as data gathering that enables you to answer your questions, and not a chance to ask them what the answers are.
- This should lead you to a list of areas to explore and the final step is to capture ways to frame these topics so that they seem spontaneous.
Some tactics
- If you have a question that you really need to ask, then find a number of paths that might produce a spontaneous “answer” before you ask the explicit question – once you ask the explicit question, you only know their answer, but you have little data to help you decide if they consider it important.
- If you have multiple people (users or customers, or other) in the conversation, then make the conversation happen between them, rather then through you. Maybe even find an excuse to leave the room for a few moments (assuming you are recording the conversations).
- If you need to offer two possibilities for their consideration, distance your self from the ideas with lines such as “In other conversations I have heard people state that they would like <x> – does that make sense to you?” “Can you explain what you think they might have meant?”
