It’s the last post of the year, and every newsletter writer in the middle of the final holiday week are reflexively reaching for the most low-risk, most highly relevant type of post for this time of year — the year end review.
But as I was thinking about the act of reflecting back upon the year, it struck me how… fascinating… the process is. After all, 365 days worth of stuff has happened in the past year, and it’s honestly impossible to remember every little detail of what has come to pass. As humans, we all suffer from recency bias (along with other memory issues) and so without an objective record to refer to, most reflections are going to be skewed in all sorts of unpredictable ways. The base dataset is just impossibly flawed, even if I refer back to “objective” sources like my newsletter post archive or photos account.
This dance of “take a quasi-objective historical record, apply human narrative of events to it, then interpret the mess to use in making decisions” is exactly what I do for a living, and yet I obviously haven’t applied any of it to private life. My life doesn’t pretend to be quasi-rational like work has to (pretends to?) be. There’s no objective function to min-max, no overarching metric to peg performance against, except perhaps to continue surviving. Anything that I happen to choose to pick out over the year is more a reflection of whatever happened to be floating in my mind at the moment than anything else.
The more I pondered upon it today, the more it it feels… weird.
At work, we’re supposed to have guard rails against just making stuff up on the fly. Sure, we still weave between surprise dumpster fires on a fairly regular basis, but there is still an bigger objective goal that we can reorient ourselves back to. This at least stops us from making every choice based off what information we can scrape together as things happen.
What’s interesting to me is that both ways of looking at the arc of our pasts in order to illuminate our future, whether casually reminiscing or more seriously picking apart past events and squeezing meaning out through analysis, life seems to work out regardless. I find this very interesting because most of us would have probably bet on the fact that the more rational would (on average) produce better results, while our completely unplanned lives that probably rely on heuristic greedy decision algos should be crashing and burning more often than they do.
So which is it? Is all that hard work we do measuring risks and rewards and running month-long A/B tests on button colors to squeeze 0.1% extra CTR on an email a complete waste of time, or is “normal life decision making” somehow fundamentally different from work decision making that justifies the differences in approaches? Because if all we’re doing to make decisions is holding up a mirror of our own organizational biases dipped in the veneer of rational analysis, I’d rather just admit this up front and save myself months of labor.
It’s less what happens before the decision, more what happens after
I still believe that if we had the time and energy to put some rational thought (and experiments!) into big decisions in life, things might turn out more optimal than they are. We just won’t ever have the energy to do so. So instead, I can think of two things that we do to make things work out.
First, we’re constantly making course corrections to all our decisions. While driving, you don’t decide to make a turn, lock your steering wheel into position, and then keep going regardless of what traffic around you is doing. You’re making changes every moment. We automatically do this in our lives without thinking twice about it — our heuristics get us close to an acceptable outcome, and our heuristics are also able to give us very quick feedback as to whether things are on a good path. We do it so often, we give it names like “muscle memory”.
I tell people who are new to industry all the time that they need to re-learn how to do this in an industry context. Decisions aren’t permanent things, and having the information and plans in place for a fast course correction is more useful. This has lots of effect on how “correct” you need to be for many decisions, and cuts down on how much rigor needs to go into doing activities.
Here’s the second thing that I think makes life “work out” despite how we don’t have time to hyper-rationalize everything. Life works out because of the hedonic treadmill, or “baseline happiness”, the gist of the concept being that most people have a baseline level of happiness that they typically return to over time, even if that person faces setbacks, tragedy, trauma, whatever, there is a tendency to bounce back. Life works out because we wind up feeling that way, even if it didn’t actually “work out” from an objective point of view.
I think this is the big difference between work and private life decision making. Work does NOT really come with a hedonic treadmill built in because work isn’t a monolithic consciousness with a base performance level. Poor decisions, such as betting $3 billion on bad real estate deals that lead to a 25% staffing layoff, have lasting effects. Making many poor choices in sequence can effectively cripple an organization for years, and employees and customers notice.
So the lesson seems to be that we are given many chances to get things eventually right, but if we don’t listen, failure does wait for us. And we can’t just “get used to it” like in private life.
Aside: There’s a phenomenon I’ve noticed but don’t have a term for, where we have a tendency to take the results of tests, even our dodgy industry ones, accept the results at face value and then move one. For example, if our A/B test says the new version is +4% improvement over control w/ statistical significance, we’ll take the winning side, and then assume we get a +4% benefit as a ratchet we keep forever going forward. But this is an analysis I’ll have to save to think more on for another day.
Things that my brain is saying is important from this year:
0 COVID infections that I know of, and 3 super high tech vaccination shots, but still have shoddy cell reception and my keys aren’t sticking to me so I keep losing them in the couch
3 sorta-viral posts, the measles-airplane post, the look at what folk were saying about Zillow, and a smaller “why’s it hard to teach data cleaning”. Since I definitely have no idea how to write a viral post on demand, I fully expect to write 0 next year and maybe accidentally stumble into one.
10 people tossed me a tip on Ko-Fi! I can’t name people, but a million thanks!!
19 people bought survivorship bias plane shirts!!
1 new homepage that’s all fancypants w/ a cloud based CI/CD setup
Finally a desktop computer upgrade after a whole year of trying to land a GPU
The very nice kitchen knife from Japan finally arrived, and then I put myself in the ER for stitches on my hand a few months later from it 🙃
I cut a bunch of gems and even sold one =O Uh, does anyone need a rupee in their life? Still haven’t learned to set things in silver yet =\
About this newsletter
I’m Randy Au, currently a Quantitative UX researcher, former data analyst, and general-purpose data and tech nerd. The Counting Stuff newsletter is a weekly data/tech blog about the less-than-sexy aspects about data science, UX research and tech. With occasional excursions into other fun topics.
All photos/drawings used are taken/created by Randy unless otherwise noted.
Curated archive of evergreen posts can be found at randyau.com
Supporting this newsletter:
This newsletter is free, share it with your friends without guilt! But if you like the content and want to send some love, here’s some options:
Tweet me - Comments and questions are always welcome, they often inspire new posts
A small one-time donation at Ko-fi - Thanks to everyone who’s supported me!!! <3
If shirts and swag are more your style there’s some here - There’s a plane w/ dots shirt available!