'Diversity' is such a woolly, fuzzy concept, because it has no proper way of being measured. Is an organisation with 99% men and 1% women diverse in sex? Does an organisation become diverse if it changes from 100% white to 99% white? Is it diversity to have a board with one person who in her or his body covers more than one ground for diversity (say, someone who is female, poor, and belongs to a racial and/or ethnic minority)?
The fuzziness allows organisations to get away with sometimes minimal changes for publicity purposes, and it also allows the concept to be exploited by powerful minorities.
Because it is not comparing the results (at least in popular debate) to population proportions of the various groups, it's harder to note that adding one white female and two black males could affect two different aspects of 'diversity' in opposite directions.
What should matter is how many of each type of demographic groups there are in a particular society (or in what is called the pipeline for particular posts or occupations in that organisation). A country where type z people are 50% should have more of them in various decision-making rules than a country where the percentage of type z people is 5%. Yet I see this often misunderstood.
So after that long tirade, because women are half of all people, diversity cannot mean that their numbers are expected to be as small as the numbers from very small demographic groups. But that is pretty much what is happening, to the detriment of women, when all 'diversity' is lumped into one big kettle and stirred like this.