Feedback: Gift or poison pill?

For the last decade, I've been deep into performance feedback. I started researching feedback in 2014, when I published a viral article in Fortune with hard numbers that showed gender bias in performance reviews. Over the last few years, I've worked with the Textio team to refresh this research annually.

Last week, we published the 2024 research. I won't recap it all here, but if you like this stuff, you should read it. The bottom line is that 1) workplace feedback continues to be biased by gender, race and age, 2) bias is starkest among high performers, and 3) people of any background who receive problematic feedback are 63% more likely to quit within the next year.

The interesting question is why none of this is getting better. After all, HR teams have spent a zillion dollars in manager training over the last decade. For this week's nerd processor, we're going to do a quick case study to understand why people still aren't good at giving fair feedback.

Except I'm not going to build the data set. You are!

Scrappy data science: DO try this at home

If you're a nerd processor regular, you know how much I love building scrappy data sets to explore social questions. In the last few months, we've built data sets to explore whether in-person or remote meetings work better, how teachers use AI, and who's likely to win the Presidential election, just to name a few.

Scrappy data science is not the same as robust research, which is what you need if you actually want to prove something. But I love scrappy data science because it gives you quick, directional signal on a hypothesis and helps you understand when a topic merits further research. You also don't need to be a professional data scientist to do this kind of analysis.

So you might be thinking that I've got a great new data set to share about the feedback gap. Well, I do -- it's in the Textio report (and in this case, it's legit research rather than scrappy data science). But for nerd processor this week, rather than presenting this data, I'm going to help you build your own scrappy data set to see why people still aren't good at giving feedback.

A fun, scrappy data set you can build for yourself

People have been talking about feedback at work for a long time. For instance, it's fun to check out some research bangers from the 1970s. In 1972, Fear of negative evaluation and the reluctance to transmit bad news, by Rosen, S., & Tesser, A., topped the charts. In 1979, Transmission of positive and negative feedback to subordinates: A laboratory investigation, by C.D. Fisher, on heavy rotation. In both academic publications and more popular writing, there's a huge body of work about feedback, and it goes back many decades.

What I find most striking about all this writing about feedback, both the scholarship and the practitioner commentary, is how it all sounds exactly the same. The 1970s stuff and the 2020s stuff and pretty much everything in between. How come?

To find out why, you're going to build your own scrappy data set.

First, search for #feedback on LinkedIn. Use the hashtag.

Next, read the first 50 posts that come back. 20 posts works, 100 is even better, but 50 is enough to be directionally interesting. As you read, classify the posts like this:

• Does the post give you a specific, concrete, and measurable action to take?
• Does it provide new, concrete data that might cause you to think differently?
• Does it share a deeply personal experience that inspires you to take action?
• Is it mostly a general platitude, like "feedback is important" or "feedback is a gift" or "people with a growth mindset FTW"

Count up your totals. For instance, here are mine:

Yes, this is scrappy data, but the trend line doesn't lie: zero posts with concrete actions, zero posts with data. A few with personal stories. 94% with general platitudes.

If you think this data set might not be watertight, you're right. It's leakier than a sieve! Maybe people don't do their real work on LinkedIn. Maybe the kind of poster who uses the #feedback hashtag is a poser. Maybe lots of things. But when the numbers show that 94% of the conversation in the world's most popular business discussion forum consists only of platitudes -- even in a platitude-heavy format like social media -- that is directionally more than random.

The bottom line: The patterns in feedback aren't changing because very few people invest in the practical, actionable tactics and tools that would cause the patterns to change. It's easier to just roll out the same ineffective trainings every year and post "#feedback is a gift" with a selfie.

The other bottom line: Data is hiding in plain sight. It's right there for you to scoop it up and count it, and you don't have to be a Data Scientist (TM) to use it.

What do you think?

Thanks for reading!

Kieran

​Catch up on nerd processor case studies | Subscribe | nerdprocessor.com​

kieran@nerdprocessor.com
​Unsubscribe · Preferences​