It's an autobiography of an economist who wanted to affect policy. This is difficult: If you want to convince a politician to do something that they didn't already want to do, you probably need some kind of rigorous argument. It's tough to set up experiments on these topics. E.g., if you want to test what happens when you raise taxes by 10%, you could randomly choose half the national population to try it and keep the other half at the same tax-rate as a control group… and then you'd get voted out of office at the next election, having infuriated half of the voters. In recent decades, economists had introduced rigor by game-theoretically showing that their proposed changes would help certain groups: e.g., assuming each hog-farmer finds out about this new hog-farming subsidy and fills in the simple 12-page form, they get some money, boosting just the right spot in the national economy, as can be mathematically shown.
In more-recent decades, behavioral economists have pointed out ways that actual human people don't behave like the theoretical-people in these economic models. The big hog farms with armies of lawyers find out about the new rules and fill in the forms; the smaller outfits never get around to it, too busy slopping hogs.
This guy, Thaler, wrote a book called Nudge about ways that real-human-people don't act like economic-model-people. And he put some ideas into practice. He studied ways to get workers to put aside money towards their retirement. He advised UK Tories on ways to get taxpayers to actually pay taxes (which I didn't even realize the Tories were in favor of, which just goes to show). Annoyingly for fans of precision, while Thaler was good at pointing out ways that economic policies for economic-model-people failed, he didn't replace that thinking with something where you could crank a formula and have exact numbers to use in a policy. It was more "Let's try doing this thing; it will almost-certainly help by some hard-to-predict fraction." Doubly-annoyingly, he was often right.
I found out about this book from a Piaw blog post from, wow, 2018, which gives you some idea of what my to-read list backlog is like these days.