Here’s a question from the mailbag:

Last week I held a presentation to my company (around 60-70 people). My strategy for the post-lunch session was to keep the audience involved by asking questions. This was to prevent people from falling asleep from the food coma.

What I did was have a slide with a question, ask the audience, see what came up, and then reveal the answer I had in mind. I got a few interesting answers and it felt like it was partially keeping people from falling to food coma. Afterwards I got some appreciative comments from programmers who attended both about the contents and about the style of the presentation.

One problem:  Our CTO thought that the way I’d asked questions when I already had the answers prepared was demeaning. He felt strongly about this in terms of “that’s something you just don’t do”.  To me, I wouldn’t be standing there holding a presentation if I didn’t think I had answers that not everyone in the room did. Asking people to think about things rather than spoon-feeding them “truth”, to me, is a way to help the process of learning.

What do you think?

There are few things “you just don’t do” as every audience is different and the rules change depending on what they’re expecting, and how good you are at using a technique.  So it’d be rare I’d ever say “you can never ever do X”.

It’s also important to realize there are many ways to reveal those answers, some which might be demeaning (“you guys were too stupid to say things like this”) but others that are entertaining, interesting and enlightening (“Here’s what I had, but your answers were better for these 2 reasons, except that you missed…”).  So how you do it is as important as what you do.

The main problem I see is whether the questions you were asking were what the audience wanted to learn.  I’d rather do one of these:

1. Ask 5 or 6 people who were going to be in the audience ahead of time what their actual questions were and use them instead of guessing.
2. Only ask them questions that reflect something I’d taught them in the lecture.  So after demonstrating how to do long division, I’d give them an easy long division problem. Then the questions tests my ability to teach as much as it tests what they know.  Any wrong answers means I failed, not the audience.

In the end the CTO is entitled to his opinion, but it’d be wise to have the opinions of the audience recorded in some simple way so his perspective is informed by what the majority felt.  Having a better way to judge the value of a session is probably the best problem to solve here.