Some thoughts on giving feedback effectively

December 16, 2014

I think a lot about the process of giving and receiving feedback a lot, because one of the concepts I'm a little obsessed with is how what you don't know, that you don't know, can have an outsized impact (the black swan theory). But also, I got a lot of feedback myself earlier in my career that I needed to improve at delivering feedback in the right forums and to the right people in a way that would actually effect the change I wanted to see.

I don't tend to have very much trouble coming up with content for feedback thanks to being trained by my Tiger Mom to always, always, relentlessly be asking and thinking about, “How could this be better? How could this be more efficient?” That incessant drive to improve is enormously powerful, but I know it can also be enormously irritating to be around, too. I've learned that appropriately managing one's more annoying tendencies can be really useful to actually getting what you want. Another result of that drive is that I often have Many Opinions and Ideas, but there can be so many of them that I don't necessarily even care all that much about a lot of them! So I've developed some tactics for achieving the long-term changes I do care most about.

I'll use the example of giving people feedback on conference talks they're practicing, since that's been something I've been doing a lot of this in the past year. One co-worker told me after one such a practice talk that he's automatically going to invite me to all his future practice talks to get my feedback, which made me feel like I might be doing something right nowadays!

There's a lot that's been written about the “compliment sandwich.” The basic idea is that because people tend to react more strongly to criticism, you should deliver positive feedback and encouragement before and after a piece of negative feedback in order to soften it. However, in practice, this can come across as kind of fake and formulaic, resulting in people still only hearing the negative feedback and then thinking you're a phony on top of it, like in this 30 Rock bit.

I think the key nugget to this idea is that you need to avoid overwhelming people with criticism. If you don't have enough substantive positive feedback of specific things people did well and should do more of, you need to prioritize and limit the pieces of negative feedback you give them, at least in that same session. You can always save the less important feedback for the next time. If I go to multiple practices of the same talk, usually at the very least, I can comment on the ways in which I observed they're put a lot of work into improving from the last practice.

When I'm sitting in on a practice talk, I take copious notes on items that probably fall into one of these categories:
  • the points that I strongly agree with, usually with some additional suggested examples for them to use to further emphasize that point
  • any slides or points I thought were particularly clever or awesome in some way
  • the parts I was confused by and any questions that came up in my mind (a preview of what might come up in Q&A)
  • my observations on any differences between the emergent structure from their talk based on my synthesis of the content as they've delivered it, and the actual structure of their talk, in order to make the talk flow more smoothly (ex. “I think part 2 was actually good background for part 1, so you should reverse the order of those two sections”)
  • any sections that could be candidates for being cut, if they're going over time
  • which sections I think they should definitely keep

The last two are particular items to watch out for, because I've noticed that in a lot of these practice talks, the presenters are usually running pretty close to time already, and then everyone (including myself) usually gives them suggestions of even more stuff they can include, which would of course bring them over time. So input on prioritizing what should be kept and what should be cut can be very useful and reduce being overwhelmed by the volume of suggested changes.

Once I have these notes down, because I tend to have a lot of them, I'll usually try to give it a moment and let other people start off with giving their feedback, while I take some time to prioritize and organize my points. Usually other people are less verbose, so it helps warm up the mood for feedback before I go in all guns blazing. The prioritization is to make sure I have enough of the positive stuff upfront and I can lead off with the suggestions I think would have the biggest impact. I also try to save a piece of really easily changed item for the end, so that if nothing else, people can seize on that small thing to change and feel good about accomplishing that (and helps me save face if they hate all my other suggestions). This prioritizing on the fly takes some practice. My notes have a lot of circles and arrows drawn on them after I do this.

Finally, another important tactic is to modulate your feedback based on how it's being received. Some people aren't very good yet at receiving feedback without feeling defensive. You can tell this is happening when you start getting a lot of explanations for why they chose to do things in that talk, versus just writing down or asking questions back to clarify the feedback. Again, because of that Tiger Mom influence, my instinct tends to be to still want to help people even if they don't really want to be helped that much right now. However, I've learned that preserving the relationship is more effective overall (and gives you a chance for your feedback to be established as trustworthy, even if it isn't taken at first), so I'll persist a little bit but then rapidly cut out a lot of what I was going to say if it seems like it isn't going over all that well.

In conclusion, this: