I am not, as far as I know, a woman, but I could find it incredibly useful to go "somewhere" and get 3-D scanned / measured and then have a giant clustered database translate my exact physical measurements into this mfgrs idea of a "M" vs this mfgrs idea of a "L" or whatever. Not to mention shoes, where I'm about a 10.75 so 50/50 odds if a 11 fits me better than a 10.5, and some kind of "big data" might help. Money saved in return shipping alone might make it financially worthwhile for amazon to run a "scanning booth" store.
I still think the best solution is to measure the normal tailors measurements with a tape, upload them, and then have manufacturers upload the exact measurements of their clothing in corresponding areas.
Crowdsourcing would allow people to mention wash shrinkage, and tolerance delta's for items.
Be cool to say I'm X size, and this is my favorite shirt. Show me more that are of similar measurements.
brb, going to start coding.
It wasn't particularly accurate, in my experience. It told me to get a medium when a small actually fit better, but it also didn't have my favorite shirt in its database, so I had to go with something that had a decent but worse fit as a comparison.