No, I haven't read Hillary Clinton's new book and I don't intend to. Five hundred and twelve pages! But I have a couple of thoughts that I haven't seen in any of the commentary I've read.
First, Clinton says she was hurt by many voters' resistance to voting for a woman. Becoming a first—first woman, first Catholic, first black—is something of problem, though it can also be an asset. Clinton certainly benefited from some voters' desire for a first woman president, though the number of voters enthused about the prospect seems to have been lower than she probably expected. Remember how in many primaries younger Democratic voters (no deplorables!) voted for Bernie Sanders.
My point is that it helps your chances of being a first if your image runs counter to negative stereotype. Margaret Thatcher wasn't a nurturing woman or a bossy feminist; she was the Iron Lady. John F. Kennedy wasn't a typical sentimental Irish pol; he had the demeanor and the wardrobe of an English lord (and in fact, one of his sisters married the Marquis of Hartington). Barack Obama, starting with his red-white-and-blue America speech at the 2004 Democratic National Convention, seemed the opposite of negative stereotypes of black people -- witness the embarrassing comments of Joe Biden and Harry Reid to precisely that effect.
Hillary Clinton, in contrast, embodied many of the negative stereotypes of liberal feminist women. On the campaign trail she flaunted her status as a woman—something which Thatcher never did. Let's just say it didn't help. My guess is that the first female American president will be more like Margaret Thatcher than Eleanor Roosevelt.
Second point: Hillary Clinton's nearly successful effort to clear the Democratic primary field ended up hurting her, and seems to be thrusting her party far to the left. It was Clinton standard operating procedure: Bill Clinton worked to dissuade other Democrats from running against Al Gore in 2000. Only Bill Bradley did so, and he bowed out after losing badly in Iowa and narrowly in New Hampshire, unable to raise enough money (surely the Clintons had something to do with this) in the multiple weeks before the next contest.
This strategy didn't work in 2016. Bernie Sanders nearly won Iowa (he may actually have had more human beings voting for him in the caucuses, but ended up just barely behind in Iowa Democrats' arcane scoring system). And Bernie whomped Clinton in New Hampshire. His seeming adherence to principle and refusal to harp on her "damned emails" endeared him to many Democratic voters, and his surprisingly good showing—43 percent of all primary and caucus votes—convinced many in the party and the press that his socialist-style politics was popular.
Imagine an alternate scenario, in which the 2016 Democratic primaries looked more like the 2016 Republican primaries, with many serious candidates running against Hillary Clinton and delegate allocation rules that favored winner-take-all over proportionate representation. She still would probably have won the nomination; even with Sanders available as an alternative for every Democrat with one reason or another to oppose her, she did win 56 percent of the primary and caucus votes. Moreover, if Democrats used something like the Republican delegate allocation systems favoring winner-take-all rather than proportional representation, she probably would have sewed up the nomination far earlier in the cycle. In the process, Sanders would have been just one of several challengers and his brand of socialism would not have netted him as many votes as it did and be as attractive as it is to many Democrats contemplating a job switch in 2020.
If you carry this alternate scenario one step further, and imagine a Republican contest with a smaller field and with delegate allocation by proportional representation, Republicans might very well not have nominated Donald Trump. But that's stuff for another blogpost.