Originally published in Computerworld.
Ever been in a situation where you wanted more information, but couldn't quite figure out how to ask the question? Whether you're eliciting requirements, interviewing vendors, interviewing candidates for a job, deciding who you want on a project team, or just asking your family about their day, you're asking questions and wanting to hear answers.
Imagine if you were interviewing a vendor and asked, “Can you walk me through an example of how that reference account implemented this software?” Or, imagine if you were a candidate for a job but your interviewer wasn't very skilled — by answering the question the interviewer should have asked, you look like a powerful and suitable candidate.
Selecting the right questions for the situation is a challenge. Questioners can choose from: closed, open-ended (context-free and behavior-description), hypothetical, meta and rhetorical questions, and auditions, to elicit the answers. Here are some techniques to choosing questions.
Closed questions establish the boundaries for the rest of your questions. Use closed questions such as “How many years of Java experience do you have?” or “How many systems have you installed vs. sold?” to establish the boundaries of what the person (or vendor) has done.
Open-ended questions explore how a person wants work performed or performs work. If you use context-free questions such as, “What does success look like?” you can understand what a senior manager wants out of this project. If you use a behavior-description question such as, “Tell me about the most challenging problem on your most recent project,” you'll learn what challenge means to a candidate.
Here's how I use context-free questions when starting a project. First, I ask, “What does success look like?” Sometimes, I hear an answer such as, “We released the software on June 15, so we could count revenue for the second quarter.” Sometimes, I hear this: “The project team is still intact.” Other times, I hear an answer such as, “Finance has the tools to do its job before the end of the year.”
I then ask
- Why are these results desirable?
- What is the solution worth?
- What problems does this system solve?
- What problems could this system create?
Armed with the resulting answers, I can define success for the project. With a definition of success, I can select a life cycle and suitable project activities, and define release criteria to create a project that has more than just a prayer of being successful.
Behavior-description questions help you differentiate the talkers from the doers. Behavior-description questions are of the form “Tell me about a time when …” or “Give me an example of …” or “On a recent project, how did you …”
If you're interviewing a vendor and want to know how easy a product installation really is, you could ask, “Please describe how the five most recent customers installed their software.” If you're interviewing a developer candidate, you could ask, “Tell me about a time when you knew you had the right design for the problem. How did you know? Did you have to convince anyone else?”
I like to also ask the converse of that question, “Tell me about a time you were stumped on a design. What did you do?” These questions help the candidate articulate how he designs and what happens when he is stuck.
Hypothetical questions help you understand how people think about a problem. For a vendor, you could ask, “How have you planned to take care of customers in such-and-so situation?” For a developer, you can ask a question such as, “What would the architecture do in this situation?” When interviewing a candidate, you can ask, “Under what circumstances would you take on a project like that again?”
Meta questions are questions about the questions. After context-free questions, I ask senior managers this meta question, “What else should I ask you?” I ask candidates, “What else should I know about you?”
Auditions help me see vendors and candidates in action, depending on the situation. When a vendor tells me the software is easy to install, I say, “Show me.” When interviewing candidates, I consider the behaviors I want to see, and then design an audition around those behaviors.
Make sure you're not asking irrelevant questions. I see these most often in interview situations. Asking a candidate where he wants to be in five years doesn't help you evaluate what a candidate can do for you now.
Match your questions to your situation, and you'll elicit the information you need.