If you ever need to present your data, and you're not ready to get a minor in graphic design, then this is the book for you! It's simple, easy to follow advice, and you can read it in an afternoon.
The most valuable message he provides is "care about presentation!", which a lot of scientists miss, arguing that they are too busy, thinking it doesn't matter, and why would you try to cater to shallow audience members anyway?
I am too new and inexperienced to say whether nice slides really make a difference in transmitting a message, but as a new and inexperienced presenter, I know that I feel waaay more confident in my presentations when my slides look nice. Also as an audience member, I can clearly remember presentations that had strikingly good slides. Therefore, I would recommend this book to any person with a lot of presentations in their future.
Pros:
The advice was overall really good, and basic enough that it doesn't require designer intuition or software to implement. Everything he shows is implemented in PowerPoint. A lot of it you could have figured out for yourself if you had actually stopped to think about it (font size, consistent color schemes), but a lot of people just don't.
I especially appreciated the example slides, and how he would present bad slides he had found, and convert them into nice ones.
Con:
As a scientist and not an economist, I think he exaggerated in his over-simplification at times. He would sometimes reduce a graph so much as to remove axis labels! This is fundamentally wrong, because anyone who got temporarily distracted should be able to look back at your plot and figure out for themselves what it shows. Also, removing that "unnecessary clutter" might make your message clearer and more convincing, but it is also cheating the public; if your effect sizes are tiny, you could completely cover that up by not providing the real axis values.