So I did. In basically four days. There was lots of unexpected plane and airport time this weekend, which was perfect for this kind of head-down, hands-on project. It was probably 12-15 hours altogether.
And it wasn't that bad. Or that hard. Just time consuming. I read some articles online as I went along and most of them told me basically what I was already doing. The index should focus on concepts, not just keywords, which is part of why software programs are insufficient. There are a lot of judgment calls involved.
I'm glad to have it. In the past few days I've also approved the back cover, revised my "About the Author" page (to help fit the index in), collected the final two blurbs, and checked all the changes made in the most recent round of line edits. The first round I had 186 and the second round 84. This final round? Just six.
(Oh, and I also put together a workshop outline for my alumni association and pitched an article to a trade publication.)
I think I am 98% done. I have the option to give it one last pass, but they're having someone new who has never read the manuscript review it, which I think is great. I may still give it a look, for typos only, but it feels good (and weird) to be really, really, really (mostly) done.