The Muselet #47: No More Blanket Feedback
One of the fun aspects of working on the ”No More” book is that I get to look at stuff I wrote some time ago, rework and sometimes completely rethink it. Here’s the result of some of this rethinking on everybody’s favorite topic of… feedback!
Where does our obsession with feedback come from?
“Please give me feedback!”
“Here’s a feedback form, please fill it in!”
“The first 10 feedback submissions get a sticker!”
And as we wait for feedback to trickle in, we read this book:
“The science and art of receiving feedback well”
“Even when it is off-base, unfair, poorly delivered, and frankly, you’re not in the mood.”
Are you kidding me!?
What are we doing to ourselves? Why is this so hard? Something is seriously wrong here.
Before we dig into the causes and look at solutions, let’s start at the source of this chaos and work our way from there. First, everybody’s favorite part: definitions!
Wikipedia defines feedback as follows:
Feedback occurs when outputs of a system are routed back as inputs as part of a chain of cause-and-effect that forms a circuit or loop. The system can then be said to feed back into itself.
The idea behind feedback is that a system (feedback comes from systems thinking) can become self-correcting and self-improving when it listens to outside signals and adjusts accordingly.
A thermostat is a good example. You set it to a target temperature, and based on sensor feedback, the thermostat decides if it should turn up the heat a bit more, or turn it down, until it reaches its target temperature.
However, something odd happened when we translated feedback to the world of the humans. It is subtly different than our thermostat example.
This is the usual “human feedback” cycle:
We do something
We ask: “got any feedback?”
The environment — often other humans — tell us: “a little bit more of this” “a bit less of that” “great job!” “booh!”
We take the feedback, process it somehow, and attempt to implement some sort of improvement, then
Can you spot the difference?
The difference is the place where the evaluation happens. As humans, when we ask for feedback, we tend to ask: what do you think? We ask for evaluation, judgment, not “sensor data.” And generally the ask is not very specific, it’s an ask for blanket feedback — gimme any feedback you got!
And that’s exactly what we get: “I love it!” “I hate it!” “That’s some pretty sloppy work you did there!”
However, is that what we really need?
Let’s look at an example. Likely , this feedback is not specifically requested, but we’re getting it anyway, let’s consider it a gift:
In that meeting you publicly disagreed with me. How dare you! Never do that again!
The judgment: disagreeing with me in public is bad bad bad. Thumbs down. No no.
This is tough, especially if this is feedback coming from your boss. Absolutely, it’s worth to listen, but perhaps it’s better not to act on (except if that boss is me, obviously). Your boss may not like your public disagreement, but is not speaking up in the company’s best interest?
The question is: what is your goal?
Hopefully, the reason you spoke up wasn’t to please your boss, but to challenge whatever the topic was at hand. For the greater good of the company.
The type of feedback you’d be interested in as a receiver is the impact your action had, not the judgement. You, in turn, will then be able to judge if this was satisfactory or not. You decide on the evaluation criteria, not your boss. Your boss not being happy being disagreed with is just one such impact signal, but perhaps the impact your disagreement had on others made it all worth it.
Alright good, I think we’re getting somewhere.
Let’s make this a bit more practical. We need to do two things:
As feedback requester: we have to take back control. Revolución!We need to take the responsibility of evaluation on ourselves rather than outsourcing it.
As feedback giver: we need to become more like sensors, skip the judgment, and provide the cleanest data we can.
Take back control
Here’s a little known fact about thermostats: even if temperature sensors have an opinion, the thermostat doesn’t care. It just needs the sensors to tell it what temperature they’re measuring. The thermostat has no interest whatsoever in any sensor’s opinion on whether these sensors feel the thermostat is doing a good job or not.
Be like the thermostat:
Decide what your goal is
Ask for specific feedback to figure out if you’re achieving it
Rather than asking after an all-hands meeting if “anybody has feedback,” ask if there is any type of information that people are missing that they would like to see covered next time.
Rather than asking if somebody likes your proposal, ask if people see any particular gaps or things you didn’t consider. And if they do, how they would improve it.
Not only will this give you more and higher quality feedback, it also will lower the chance of feeling judged in the process.
No more blanket feedback!
Based on what you get back, you can do your own evaluation. You can judge yourself if you’re on the right track. And as it turns out, judging yourself is far less painful than being judged by others.
Why do we need books be written about receiving feedback? Because receiving judgment from others often feels like a personal attack. Perhaps it shouldn’t, but hey — we are human. Yet, feedback is a gift, so we need to fight human nature: we need to learn to smile, nod, say “thank you, more please!” while our attacker keeps pounding on our face.
No more judgmental feedback.
Even if we try to get specific, sensorial data, there is a challenge: people make for pretty damn bad sensors. People are more irrational than we would like, and even if we try to ask for clean signal, we usually still get some sort of biased judgment.
Just for fun, have a browse through the list of cognitive biases on wikipedia. There are dozens, maybe hundreds of them. Anchoring, confirmation bias, framing effect, gender biases, culture biases, projection bias, halo effect, groupthink, recency bias. The list goes on and on.
I love people, don’t get me wrong, but people make for very bad sensors, so if there’s any alternative way to gather the data we need to evaluate our level of success, let’s use it.
Practically speaking, in many situations we cannot realistically do without feedback from human beings. Especially in the management line of business.
Giving feedback, nonviolently
Now let’s switch sides, let’s say we are asked to give feedback. Hopefully, it was in the format suggested in the previous section. If so, this should be relatively straight forward.
“In the all hands, you only call out sales successes. As a person in engineering, I feel excluded. Can we include more information on what engineering has been up to?” Great.
However, chances are that the feedback requester hasn’t read this email, so they will just ask for blanket feedback, “gimme some!” Or… perhaps you weren’t specifically asked for feedback, but you feel the urge to give some anyway.
How do we approach this in the best way possible?
The general approach would be: be like sensors, share observations without judgment, without the violence.
Describe the situation/action in question
Describe the impact it had on you, others, the environment
“When you disagreed with me in that meeting (1), I got angry because I felt it undermined my authority (2).”
“Ever since you gave that talk about end-to-end testing (1), I’ve noticed many more engineers add end-to-end tests to their code (2).”
“Whenever you lean back eating your sandwich (1), mom and dad have a lot more bread crumbs to clean from the floor (2).”
“When you called me an doody head (1) it made me feel teary tear (2).”
No judgement, just observations and impact.
On the receiving end: no defensive response expected. No need to buy a box of books for your team on the art of “receiving feedback well” in preparation of your 1:1 meetings.
Observation and impact, that’s all. It’s for the receiving end to figure out how to use this signal, to judge for themselves if the impact is desired or not and adjust accordingly.
Or ignore your feedback entirely. You’re probably biased anyway.