Custom bras can't be automated-- they're too difficult
December 25, 2019 5:41 AM Subscribe
You can't automate something that isn't well understood, and making custom bras isn't well understood. The article includes a description of the difficulties the startup faced (a lot of variation in breast and body shape) and advice on where to find bras.
Link from Andrew Ducker, who does a daily collection of interesting links.
Link from Andrew Ducker, who does a daily collection of interesting links.
I find it interesting that there's no hand-crafted legacy to draw on, unlike shirts, etc. The problem seems to boil down to:
1) Model what you have ( Are you using the correct model and parameters?)
2) Model how you want what you have to look like. ( Are you using the correct model and parameters?)
3) Calculate the "function" to transform #1 to #2 ( Show all work. Use additional sheets as needed )
4) Implements #3 in fabric.
Credit where due. They took on a big problem and swung for the fences. Apparently, as GenjiandProust commented, they chose the wrong tool, and/or didn't use it right trying to train an AI algorithm when their models might not have been fully baked.
It appears that the primary factor to getting a well fitted bra will continue to be trying them on until you get lucky, although the experience of a professional sales staff may take a lot of the trial-and-error out of the process.
posted by mikelieman at 6:23 AM on December 25, 2019 [4 favorites]
1) Model what you have ( Are you using the correct model and parameters?)
2) Model how you want what you have to look like. ( Are you using the correct model and parameters?)
3) Calculate the "function" to transform #1 to #2 ( Show all work. Use additional sheets as needed )
4) Implements #3 in fabric.
Credit where due. They took on a big problem and swung for the fences. Apparently, as GenjiandProust commented, they chose the wrong tool, and/or didn't use it right trying to train an AI algorithm when their models might not have been fully baked.
It appears that the primary factor to getting a well fitted bra will continue to be trying them on until you get lucky, although the experience of a professional sales staff may take a lot of the trial-and-error out of the process.
posted by mikelieman at 6:23 AM on December 25, 2019 [4 favorites]
Not to abuse the edit window, in conclusion, fuck the patriarchy.
posted by mikelieman at 6:24 AM on December 25, 2019 [16 favorites]
posted by mikelieman at 6:24 AM on December 25, 2019 [16 favorites]
I guess it’s nice that they had the candor to write a post-mortem, but what the fuck seriously. That diagram alone is enough to tell you they know nothing about tits. There are five “problem” variations in boobs? Nooo. There are many many variations in boobs that definitely can’t be accounted for in a cup size and a band size.
I really don’t think they’re speaking in good faith since they talk about the history of men’s suits and the industrial revolution and shit. It doesn’t make sense. Information about sizing a bra properly is very much available in the year 2019 and if you want to start a bra company without learning about that maybe... I don’t know. You’re fucked. You’re gonna fold.
posted by Sterros at 6:42 AM on December 25, 2019 [28 favorites]
I really don’t think they’re speaking in good faith since they talk about the history of men’s suits and the industrial revolution and shit. It doesn’t make sense. Information about sizing a bra properly is very much available in the year 2019 and if you want to start a bra company without learning about that maybe... I don’t know. You’re fucked. You’re gonna fold.
posted by Sterros at 6:42 AM on December 25, 2019 [28 favorites]
While tailors have figured out a formula for men’s suits, bra tailoring is a younger technology with a smaller market
Record scratch, roughly 50% of the clothing market is men who might buy suits, and 50% of the market is women who might buy bras. How is the bra-buying market “smaller”?
posted by saucysault at 6:52 AM on December 25, 2019 [36 favorites]
Record scratch, roughly 50% of the clothing market is men who might buy suits, and 50% of the market is women who might buy bras. How is the bra-buying market “smaller”?
posted by saucysault at 6:52 AM on December 25, 2019 [36 favorites]
We could not automate that which we did not know how to do by hand.
I feel like this is an extremely common blind spot in the tech sector, and possibly in most of western consumer culture. There is often a tremendous amount of skill and experience that goes into creating goods we consider to be “mass produced”, but that work is done by people we can’t or choose not to see.
posted by mhoye at 6:55 AM on December 25, 2019 [83 favorites]
I feel like this is an extremely common blind spot in the tech sector, and possibly in most of western consumer culture. There is often a tremendous amount of skill and experience that goes into creating goods we consider to be “mass produced”, but that work is done by people we can’t or choose not to see.
posted by mhoye at 6:55 AM on December 25, 2019 [83 favorites]
They're talking about custom, one-off, made-to-fit bras. Most women are not in the market for those. I have difficult to fit breasts and even I don't see the value in $135-$200 custom-fit bras.
posted by muddgirl at 7:03 AM on December 25, 2019 [15 favorites]
posted by muddgirl at 7:03 AM on December 25, 2019 [15 favorites]
Record scratch, roughly 50% of the clothing market is men who might buy suits, and 50% of the market is women who might buy bras. How is the bra-buying market “smaller”?
They go into details about the market size later -- it's not referring to the pure market in people who might conceivably buy the thing, but the actual amount of money that might be spent on a particular product:
Ultimately, a custom-made bra is still going to cost $135, which accounts for 2% of the U.S. lingerie market. At the end of the day, this is a niche market, and for those who can afford a luxury item.Our conservative calculation of the market size results in $55 million.
posted by jacquilynne at 7:05 AM on December 25, 2019 [3 favorites]
They go into details about the market size later -- it's not referring to the pure market in people who might conceivably buy the thing, but the actual amount of money that might be spent on a particular product:
Ultimately, a custom-made bra is still going to cost $135, which accounts for 2% of the U.S. lingerie market. At the end of the day, this is a niche market, and for those who can afford a luxury item.Our conservative calculation of the market size results in $55 million.
posted by jacquilynne at 7:05 AM on December 25, 2019 [3 favorites]
I wish people would actually read the article before coming in to give us their hot takes on AI and the patriarchy. This seems a genuinely difficult problem that they underestimated and the article does lay out the issues in a thoughtful way.
posted by peacheater at 7:09 AM on December 25, 2019 [40 favorites]
posted by peacheater at 7:09 AM on December 25, 2019 [40 favorites]
I did read the article. They aren’t comparing suits for “hard to fit” people and bras for “hard to fit” breast-bearing people. They were comparing the market for ALL suits vs ALL bras.
posted by saucysault at 7:16 AM on December 25, 2019 [3 favorites]
posted by saucysault at 7:16 AM on December 25, 2019 [3 favorites]
They were discussing bra tailoring, which they assert (believably, imho) to be a measurably smaller market. I think they could ultimately move the needle on the market size if successful, though: if they can mass-produce custom fit bras, they should eventually be able to lower the price and grow the market.
Honestly, they probably started from a similar point of outrage: Why is it so hard to get this ubiquitous piece of clothing that LargePercentage of the population wear every day?
As it turns out, there's less pre-existing knowledge, and the problem itself is pretty different from suit tailoring. At the risk of making an ignorant misunderstanding, the 'long tail' of suit tailoring mostly comes down to 'make it bigger/smaller, and deal with the smaller market size.' IIUC, suits give you a geometry problem, and bras give you a physics problem. So in the bra business, they were hitting harder design problems, and had more axes of important variation to deal with.
In the write-up, it was clear to me that the body type variations were their design problems to solve, not problems with the customers, if that makes sense. They're not saying 'these bodies are wrong,' they're saying 'these are the bodies we failed to serve.' And by sharing the information this way, they might help someone else down the line.
posted by kaibutsu at 7:33 AM on December 25, 2019 [27 favorites]
Honestly, they probably started from a similar point of outrage: Why is it so hard to get this ubiquitous piece of clothing that LargePercentage of the population wear every day?
As it turns out, there's less pre-existing knowledge, and the problem itself is pretty different from suit tailoring. At the risk of making an ignorant misunderstanding, the 'long tail' of suit tailoring mostly comes down to 'make it bigger/smaller, and deal with the smaller market size.' IIUC, suits give you a geometry problem, and bras give you a physics problem. So in the bra business, they were hitting harder design problems, and had more axes of important variation to deal with.
In the write-up, it was clear to me that the body type variations were their design problems to solve, not problems with the customers, if that makes sense. They're not saying 'these bodies are wrong,' they're saying 'these are the bodies we failed to serve.' And by sharing the information this way, they might help someone else down the line.
posted by kaibutsu at 7:33 AM on December 25, 2019 [27 favorites]
They aren’t comparing suits for “hard to fit” people and bras for “hard to fit” breast-bearing people. They were comparing the market for ALL suits vs ALL bras.I mean, look, I am basically on your side on this one. I have the dreaded small-back, big-boob issue, and finding bras has been a constant source of drama for me since I hit puberty at the age of 11. There is no place within a hundred miles of my house that sells a bra in my size, and I truly, truly blame the patriarchy for that. But this is a business, dealing with the actual realities of operating a business. And the fact is, rich men have been conditioned for generations to believe that a custom suit is a status marker, and women have spent generations being conditioned to make due with the best bras they can find. I pay *a lot* for bras, but I'm not in the market for a $135 custom bra. And if I'm not, then probably not a lot of people are. There are all sorts of patriarchy-related reasons that people won't buy expensive bras, but that doesn't make any difference if you're trying to make a go of your business, unless you have a realistic reason to think you can change that in a pretty quick timeframe.
posted by ArbitraryAndCapricious at 7:33 AM on December 25, 2019 [19 favorites]
It seems like maybe the measurements they were taking, or having customers take, and feeding into the algo weren't well-aligned to the outputs they needed in order to actually make a bra?
I've seen this happen with other attempts to use AI/ML (or just sophisticated algorithms of other flavors) to solve problems: if the inputs and outputs aren't correlated, you're going to have a bad time. You can feed the model all the data you want, collected as carefully and with as much precision as you like, and it won't help you.
With a traditional tailored suit, the measurements that are being taken are basically the outputs. To determine the length of a sleeve, you measure the length of the wearer's arm, down to where you want the cuff to fall. As algorithms go, it's pretty simplistic. (I'd be willing to bet that using something like a convolutional neural network would actually get you worse results than a simple linear regression between input measurements and outputs, for most aspects of a suit.) And even then, it's pretty rare for a tailored suit to be perfect on the first try, generally there's a second or even third fitting, basically iterating until it fits well enough to satisfy the customer. Bras seem like they'd be hard to do that with, at least economically.
It looks like Bra Theory tried to take on a really hard problem and made an honest try at it, which is always painful to see not succeed. I wonder if they will open source their algorithm and any other artifacts—it seems like they definitely moved the state of the art forward, just not enough to be commercially successful.
posted by Kadin2048 at 8:02 AM on December 25, 2019 [8 favorites]
I've seen this happen with other attempts to use AI/ML (or just sophisticated algorithms of other flavors) to solve problems: if the inputs and outputs aren't correlated, you're going to have a bad time. You can feed the model all the data you want, collected as carefully and with as much precision as you like, and it won't help you.
With a traditional tailored suit, the measurements that are being taken are basically the outputs. To determine the length of a sleeve, you measure the length of the wearer's arm, down to where you want the cuff to fall. As algorithms go, it's pretty simplistic. (I'd be willing to bet that using something like a convolutional neural network would actually get you worse results than a simple linear regression between input measurements and outputs, for most aspects of a suit.) And even then, it's pretty rare for a tailored suit to be perfect on the first try, generally there's a second or even third fitting, basically iterating until it fits well enough to satisfy the customer. Bras seem like they'd be hard to do that with, at least economically.
It looks like Bra Theory tried to take on a really hard problem and made an honest try at it, which is always painful to see not succeed. I wonder if they will open source their algorithm and any other artifacts—it seems like they definitely moved the state of the art forward, just not enough to be commercially successful.
posted by Kadin2048 at 8:02 AM on December 25, 2019 [8 favorites]
I wonder if they will open source their algorithm and any other artifacts—it seems like they definitely moved the state of the art forward, just not enough to be commercially successful.
I think this highlights some of the inherent limitations in the ability of for-profit enterprises to do research: because their profit model basically depends on them creating secret knowledge, first they can't really talk to anyone else while they're doing the research, and then if their research doesn't become profitable in just a few years they have to abandon it, and finally the abandoned research is very rarely published. So nobody can collaborate, duplicate data is collected at great expense and never pooled, and everyone repeats the same mistakes.
posted by Pyry at 8:15 AM on December 25, 2019 [27 favorites]
I think this highlights some of the inherent limitations in the ability of for-profit enterprises to do research: because their profit model basically depends on them creating secret knowledge, first they can't really talk to anyone else while they're doing the research, and then if their research doesn't become profitable in just a few years they have to abandon it, and finally the abandoned research is very rarely published. So nobody can collaborate, duplicate data is collected at great expense and never pooled, and everyone repeats the same mistakes.
posted by Pyry at 8:15 AM on December 25, 2019 [27 favorites]
But, but, AI can do anything!
They note in the link they were not able to use AI ("machine learning") approaches due to the cost of collecting enough data to use that approach. (With ML, you would need lots and lots of data points — fittings of test designs — to determine what variables should go into creating custom, de novo designs. This could include a variety of measurements and fit criteria that make data collection and usage difficult and expensive.) That does make me wonder how they thought they could go in more or less cold, into automating a customized product without some domain knowledge.
posted by They sucked his brains out! at 8:52 AM on December 25, 2019 [1 favorite]
They note in the link they were not able to use AI ("machine learning") approaches due to the cost of collecting enough data to use that approach. (With ML, you would need lots and lots of data points — fittings of test designs — to determine what variables should go into creating custom, de novo designs. This could include a variety of measurements and fit criteria that make data collection and usage difficult and expensive.) That does make me wonder how they thought they could go in more or less cold, into automating a customized product without some domain knowledge.
posted by They sucked his brains out! at 8:52 AM on December 25, 2019 [1 favorite]
The measuring part was what I was most interested in. Because the size/shape of a desidered bra if different to the size/shape of the breast (unless you still have perky ones). And different breast may measure the same when pendulous, but due to the density of the breast itself would require a different shaped bra. It is an interesting mathematical problem - good on the founder for trying to solve a mostly unnoticed problem affecting so many people.
posted by saucysault at 9:01 AM on December 25, 2019 [2 favorites]
posted by saucysault at 9:01 AM on December 25, 2019 [2 favorites]
Shout out to Mary Phelps Jacob, the woman who invented the modern bra and then went on to become one of the most influential publishers and arts patrons of the twentieth century.
posted by interogative mood at 10:58 AM on December 25, 2019 [16 favorites]
posted by interogative mood at 10:58 AM on December 25, 2019 [16 favorites]
The Jacobs biography is amazing and describes passim how the materials and design goals of bras keep changing. Millimeter accuracy needed in asymmetric curve-to-curve seams in slithery stretch fabric for one of the squishier parts of the body and a moving fashion target is a TERRIBLE problem.
Bringing together professional custom bra-makers, and mathematicians, and pattern makers (different job), is as thorough an approach as I’ve ever heard of. Big bra companies try new fitting approaches and fail regularly. The books on making your own are detailed and don’t all agree with each other. The compromise between looks and comfort is subjective. It’s really, really difficult.
posted by clew at 11:34 AM on December 25, 2019 [16 favorites]
Bringing together professional custom bra-makers, and mathematicians, and pattern makers (different job), is as thorough an approach as I’ve ever heard of. Big bra companies try new fitting approaches and fail regularly. The books on making your own are detailed and don’t all agree with each other. The compromise between looks and comfort is subjective. It’s really, really difficult.
posted by clew at 11:34 AM on December 25, 2019 [16 favorites]
I thought this was a great and thorough reckoning of their failure. I am the target market for $135 custom bras but they’ve all been let downs (uhh...pun intended).
I have a small back relative to my cup size but I’m also fat and my squishy side and back means that in order to get a lot of support out of the band I have to put up with deep welts. According to A Bra That Fits that is normal and okay. I was measured by one of the two stores in Portland (which is a lot!) that specializes in hard to fit sizes and got a bra that yes, “fits” perfectly. It gives me fantastic cleavage. However it’s unbearably painful to wear and I regret the $90 I spent on it, but better than the bra I’ve spent double that on that was custom and had the same issue.
The bras I wear everyday now are by Evelyn & Bobbie, a local Portland company. They are as supportive as underwire and give me a smooth silhouette. They’re miraculous. But the reason I ended up with one in the first place is because they used one of my friends as a fit model and they also had her do the thing where they attached sensors to her and had to run and jump and bend over so they could design a better bra with this information. Despite her being literally the customer they tried to design for, she ended up hating the bra they gave her for her troubles and gave it to me. And weirdly enough she’s 50-75lbs lighter than me and is at least 3 cup sizes smaller but it fits me perfectly. In fact the one she gave me has stretched out so now it’s a bit too big. According to their size chart I should wear their largest size but my next purchase from them will be two sizes down from that.
In summary, this is really fucking hard to get right.
posted by the thorn bushes have roses at 12:04 PM on December 25, 2019 [29 favorites]
I have a small back relative to my cup size but I’m also fat and my squishy side and back means that in order to get a lot of support out of the band I have to put up with deep welts. According to A Bra That Fits that is normal and okay. I was measured by one of the two stores in Portland (which is a lot!) that specializes in hard to fit sizes and got a bra that yes, “fits” perfectly. It gives me fantastic cleavage. However it’s unbearably painful to wear and I regret the $90 I spent on it, but better than the bra I’ve spent double that on that was custom and had the same issue.
The bras I wear everyday now are by Evelyn & Bobbie, a local Portland company. They are as supportive as underwire and give me a smooth silhouette. They’re miraculous. But the reason I ended up with one in the first place is because they used one of my friends as a fit model and they also had her do the thing where they attached sensors to her and had to run and jump and bend over so they could design a better bra with this information. Despite her being literally the customer they tried to design for, she ended up hating the bra they gave her for her troubles and gave it to me. And weirdly enough she’s 50-75lbs lighter than me and is at least 3 cup sizes smaller but it fits me perfectly. In fact the one she gave me has stretched out so now it’s a bit too big. According to their size chart I should wear their largest size but my next purchase from them will be two sizes down from that.
In summary, this is really fucking hard to get right.
posted by the thorn bushes have roses at 12:04 PM on December 25, 2019 [29 favorites]
Interesting. I have hard to fit boobs, really the boobs are fine, its the way they are attached thats the problem. And I'm tall and tall bras are rarer than hens teeth. Bra fitting is really complicated, much more so than jackets. There are so many options and there is a reason lace up supporters were so popular for so long.
I could see paying for this. If I needed to wear a traditional bra often, especially if I often needed one for formal wear or specific outfit types. Models, actresses, corporate types etc probably would happily pay for a service that allowed them to own a few custom bras. I might buy on thta went well under sleveless tops and shells, which is my hardest to find bra style. It would save me money in the long run
For now though, I've switched almost completely to Kinflyte bra tops as the best alternative for every day wear. Not sure how they'll work with tank tops in the summer tho.
posted by fshgrl at 12:08 PM on December 25, 2019 [3 favorites]
I could see paying for this. If I needed to wear a traditional bra often, especially if I often needed one for formal wear or specific outfit types. Models, actresses, corporate types etc probably would happily pay for a service that allowed them to own a few custom bras. I might buy on thta went well under sleveless tops and shells, which is my hardest to find bra style. It would save me money in the long run
For now though, I've switched almost completely to Kinflyte bra tops as the best alternative for every day wear. Not sure how they'll work with tank tops in the summer tho.
posted by fshgrl at 12:08 PM on December 25, 2019 [3 favorites]
Not to abuse the edit window, in conclusion, fuck the patriarchy.
Thats idiotic. Bras are not a tool of oppression, they allow me to do high impact sports and be more comfortable day to day than I would be otherwise. And yes I wear regular bras for sports mostly as they are more supportive than most aports bras and accomodate tall people better.
I'm glad to see more investment in good fitting bras but there has been plenty in the past and will continue to be going forward becuase women want and will pay for it and always have.
posted by fshgrl at 12:25 PM on December 25, 2019 [10 favorites]
Thats idiotic. Bras are not a tool of oppression, they allow me to do high impact sports and be more comfortable day to day than I would be otherwise. And yes I wear regular bras for sports mostly as they are more supportive than most aports bras and accomodate tall people better.
I'm glad to see more investment in good fitting bras but there has been plenty in the past and will continue to be going forward becuase women want and will pay for it and always have.
posted by fshgrl at 12:25 PM on December 25, 2019 [10 favorites]
I think that was less "fuck the patriarchy, bras suck" and more "fuck the patriarchy, why are bras still not any better?" which as a tall person I fully agree with. If men had boobs, we'd have a billion options, all affordable, and a well understood and functional standard approach to bra sizing and fitting, as well as widespread and generally excellent tailoring services. Instead we get a half-assed system of just band and cup that kinda sorta maybeworks if you're near enough average, and god help you otherwise.
It's not that bras are a bad thing, it's that the bras we have aren't as good as they can and should be (at least for most people).
posted by Dysk at 12:35 PM on December 25, 2019 [21 favorites]
It's not that bras are a bad thing, it's that the bras we have aren't as good as they can and should be (at least for most people).
posted by Dysk at 12:35 PM on December 25, 2019 [21 favorites]
It seems like maybe the measurements they were taking, or having customers take, and feeding into the algo weren't well-aligned to the outputs they needed in order to actually make a bra?
What if the data inputs needed are more difficult to quantify than expected? What if they needed to put people on treadmills wearing bras made of materials of various degrees of elasticity and measure results like Dr Michelle Norris at University of Portsmouth? I expect you have to restrict the data to information that's commercially viable to obtain. If you want some pure Internet play you aren't going to do something like what the insole companies do with the machines you stand on in the store that measure pressure at the sole. Especially if you only have 20! people to work with initially. This may not be a solvable problem even if they'd had the vision and charisma to stand up in front of VC's and believably articulate something like If we solve this we'll take not just the custom bra market but create a technology that will dominate every clothes market.
I love the fact that Mona Zhang did a write up on lessons learned, and that was definitely one of the best analytical write ups I've seen for what's got to be an emotionally gutting decision. Looking at her other writing it shouldn't be a surprise.
Maybe someday someone will come at the same problem from a slightly different direction like improving athletic performance. Not that comfort isn't important, but people are willing to pay an enormous amount of money for things that might give a minuscule performance advantage whereas an extra $10 for something that might feel better or chafe less is a tough sell. Especially if they've already been burned by buying custom fit bras that haven't worked out. I used to say when I worked for a startup that we were pitching to VCs haunted by the ghosts and charred wreckage of prior companies efforts. You want the smart money backing you, people who've been in the space you are investigating, but when they failed at the same kind of endeavor you have to be a genius to pitch to them without singing their ego. Why do you think you can do better at this than I did? Custom clothing probably doesn't have that issue as much with VCs, but I'm sure the buying public who is the potential market is skeptical at best.
posted by BrotherCaine at 12:38 PM on December 25, 2019 [4 favorites]
What if the data inputs needed are more difficult to quantify than expected? What if they needed to put people on treadmills wearing bras made of materials of various degrees of elasticity and measure results like Dr Michelle Norris at University of Portsmouth? I expect you have to restrict the data to information that's commercially viable to obtain. If you want some pure Internet play you aren't going to do something like what the insole companies do with the machines you stand on in the store that measure pressure at the sole. Especially if you only have 20! people to work with initially. This may not be a solvable problem even if they'd had the vision and charisma to stand up in front of VC's and believably articulate something like If we solve this we'll take not just the custom bra market but create a technology that will dominate every clothes market.
I love the fact that Mona Zhang did a write up on lessons learned, and that was definitely one of the best analytical write ups I've seen for what's got to be an emotionally gutting decision. Looking at her other writing it shouldn't be a surprise.
Maybe someday someone will come at the same problem from a slightly different direction like improving athletic performance. Not that comfort isn't important, but people are willing to pay an enormous amount of money for things that might give a minuscule performance advantage whereas an extra $10 for something that might feel better or chafe less is a tough sell. Especially if they've already been burned by buying custom fit bras that haven't worked out. I used to say when I worked for a startup that we were pitching to VCs haunted by the ghosts and charred wreckage of prior companies efforts. You want the smart money backing you, people who've been in the space you are investigating, but when they failed at the same kind of endeavor you have to be a genius to pitch to them without singing their ego. Why do you think you can do better at this than I did? Custom clothing probably doesn't have that issue as much with VCs, but I'm sure the buying public who is the potential market is skeptical at best.
posted by BrotherCaine at 12:38 PM on December 25, 2019 [4 favorites]
So George Lucas was right and there won't be any space bras in the future
posted by glasseyes at 1:06 PM on December 25, 2019 [2 favorites]
posted by glasseyes at 1:06 PM on December 25, 2019 [2 favorites]
but what the fuck seriously?
Yeah, what the fuck? They spent like a year doing some bullshit before trying to make 25 bespoke bras, and then they immediately folded because they figured out they didn’t know shit about actually making bras? Who gave these assholes money?
Also, I had to double check the article to make sure it wasn’t about ThirdLove since the narrative (we tried automating bra sizing using technology, but eventually said ‘fuck it’ because it’s too hard) is practically the same. The big difference is ThirdLove abandoned ship on the technology side back in 2015 and went to the tried-and-true biz model of “try a bunch and send back the ones that don’t work”.
Also, ThirdLove is doing just under a $100M a year, so not sure where the $55M number in the article came from. Unless they are trying to No True Scotsmen the market to soften their failure.
posted by sideshow at 1:09 PM on December 25, 2019 [2 favorites]
Yeah, what the fuck? They spent like a year doing some bullshit before trying to make 25 bespoke bras, and then they immediately folded because they figured out they didn’t know shit about actually making bras? Who gave these assholes money?
Also, I had to double check the article to make sure it wasn’t about ThirdLove since the narrative (we tried automating bra sizing using technology, but eventually said ‘fuck it’ because it’s too hard) is practically the same. The big difference is ThirdLove abandoned ship on the technology side back in 2015 and went to the tried-and-true biz model of “try a bunch and send back the ones that don’t work”.
Also, ThirdLove is doing just under a $100M a year, so not sure where the $55M number in the article came from. Unless they are trying to No True Scotsmen the market to soften their failure.
posted by sideshow at 1:09 PM on December 25, 2019 [2 favorites]
Also, ThirdLove is doing just under a $100M a year, so not sure where the $55M number in the article came from.
ThirdLove aren't tailoring bras. You buy preexisting bras off the shelf. That's not a no true Scotsman difference so much as pointing out that a Welshman isn't from Scotland.
posted by Dysk at 1:28 PM on December 25, 2019 [3 favorites]
ThirdLove aren't tailoring bras. You buy preexisting bras off the shelf. That's not a no true Scotsman difference so much as pointing out that a Welshman isn't from Scotland.
posted by Dysk at 1:28 PM on December 25, 2019 [3 favorites]
Relieved to learn that ThirdLove isn't a website for third cousin dating. Because that's my billion dollar startup idea. If you've got $10 million to invest, let's talk.
posted by bunbury at 1:36 PM on December 25, 2019 [8 favorites]
posted by bunbury at 1:36 PM on December 25, 2019 [8 favorites]
True & Co. makes some cuts to (I think) fill gaps they found in existing offerings - their house brand is Bezier!
I’m not sure a world in which breast-havers were the assumed dominant class would have perfected bras like the current ideal bra, suitable for the non dominant class. I think it’s as likely that fashion would attend to the shapes of rich, middle aged breasts and not expect too much discomfort from the rest of them.
posted by clew at 1:43 PM on December 25, 2019 [2 favorites]
I’m not sure a world in which breast-havers were the assumed dominant class would have perfected bras like the current ideal bra, suitable for the non dominant class. I think it’s as likely that fashion would attend to the shapes of rich, middle aged breasts and not expect too much discomfort from the rest of them.
posted by clew at 1:43 PM on December 25, 2019 [2 favorites]
I suggest that the tech just isn't there yet, and note that the vast majority of men can't afford expertly tailored suits.
Maybe in ten or twenty years computer programs, neurology and material science will have advanced to the point where making custom bras can be automated.
posted by Nancy Lebovitz at 1:58 PM on December 25, 2019
Maybe in ten or twenty years computer programs, neurology and material science will have advanced to the point where making custom bras can be automated.
posted by Nancy Lebovitz at 1:58 PM on December 25, 2019
the vast majority of men can't afford expertly tailored suits.
They also don't need to, because off the shelf suits fit the vast majority of men just fine, and there are simple measurements that enable you to barely have to try a given one on top know that it will fit well. The tailoring is a red herring here in a sense.
posted by Dysk at 2:16 PM on December 25, 2019 [7 favorites]
They also don't need to, because off the shelf suits fit the vast majority of men just fine, and there are simple measurements that enable you to barely have to try a given one on top know that it will fit well. The tailoring is a red herring here in a sense.
posted by Dysk at 2:16 PM on December 25, 2019 [7 favorites]
As someone who has had to buy both suits and bras: the former is a lot easier, there are many more options to cover much more variety in body shape and size, and for a given set of measurements, there is much greater consistency across both individual garments and different brands.
posted by Dysk at 2:18 PM on December 25, 2019 [1 favorite]
posted by Dysk at 2:18 PM on December 25, 2019 [1 favorite]
From the article:
I switched from bras to binders a few years ago, and the biggest surprise was that given a decently made binder, binding is actually more comfortable than wearing a bra. Holding the weight in close to the body means better weight distribution, which means easier breathing and less strain on my back and shoulders. Physics!
posted by Basil Stag Hare at 3:19 PM on December 25, 2019 [15 favorites]
The more pendulous the breast, the more difficult it was to relate measurements on the body to measurements of the pattern.This is the central difficulty of bra engineering, and the point at which the analogy to custom suits, etc, breaks down. Even the best tailoring can't get past the fact that you're taking a body part which is one shape and putting it into a garment that is supposed to make it a completely different shape and expecting that to be comfortable and pleasant to wear for long periods.
I switched from bras to binders a few years ago, and the biggest surprise was that given a decently made binder, binding is actually more comfortable than wearing a bra. Holding the weight in close to the body means better weight distribution, which means easier breathing and less strain on my back and shoulders. Physics!
posted by Basil Stag Hare at 3:19 PM on December 25, 2019 [15 favorites]
It's wild to me that they approached this problem without bothering to collect a lot of data first. It seems to me the first thing you'd do is find a lot of bra-wearers of various body types, take detailed measurements of and around their breasts, and then also take a bunch of measurements of their favorite bras to see how the bra and breast measurements compare, then ask them how they feel about the fit of their bra. Get as many sets of this information as possible, I'd say at least a few thousand to start.
Then, with that, you could start to build a model for how bras and breasts and bodies interact and how to predict what kind of bra will feel comfortable based on someone's breast measurements. It absolutely seems like a huge problem and one worth solving, but these folks seemed to only be paying the lightest lip service to using anything beyond the guesswork that normal bra buying involves.
posted by Mr.Encyclopedia at 5:59 PM on December 25, 2019 [3 favorites]
Then, with that, you could start to build a model for how bras and breasts and bodies interact and how to predict what kind of bra will feel comfortable based on someone's breast measurements. It absolutely seems like a huge problem and one worth solving, but these folks seemed to only be paying the lightest lip service to using anything beyond the guesswork that normal bra buying involves.
posted by Mr.Encyclopedia at 5:59 PM on December 25, 2019 [3 favorites]
Yeah bras work best on perky boobs and thinner people. Lace up bodices and binders fit a much wider variety of boobage. Its just physics.
The Bra That Fits people are nuts imho with the band sizes. Ita not healthy to wear such a tight garment. I wear a full 3-4" larger band than they recommmend and it works fine. Then again, I don't need lift as much as containment. If I needed lift I think I'd switch to bodice/corset styles personally.
Btw Third Love is just selling a bra you used to be able to buy on amazon for $20 for years before that company started. I still have an old one somewhere and its identical.
posted by fshgrl at 6:02 PM on December 25, 2019
The Bra That Fits people are nuts imho with the band sizes. Ita not healthy to wear such a tight garment. I wear a full 3-4" larger band than they recommmend and it works fine. Then again, I don't need lift as much as containment. If I needed lift I think I'd switch to bodice/corset styles personally.
Btw Third Love is just selling a bra you used to be able to buy on amazon for $20 for years before that company started. I still have an old one somewhere and its identical.
posted by fshgrl at 6:02 PM on December 25, 2019
Wanted to add that the bodice style seems to be making a semi comeback with the long line bra trend. I have one with very minimal structure and its far more comfortable than a regular bra. No pressure on the shoulders hardly at all and freedom of the ribcage too. I've hiked in it and loved it.
posted by fshgrl at 6:18 PM on December 25, 2019 [2 favorites]
posted by fshgrl at 6:18 PM on December 25, 2019 [2 favorites]
> It's wild to me that they approached this problem without bothering to collect a lot of data first.
Even if we hand-wave a magic algorithms could solve the problem, they just didn't have enough data. Nor do I see a way to get the level of data I think would be needed. "Please send us comprehensive photos of your breasts (including some with hands cradling them) . Additionally, we'll need video of them while you jump around, in order for us to create a perfect bra. We'll also need your name and shipping address." Uh huh. (Additionally, I don't believe pictures are sufficient enough, since the internal make-up varies between breast, fat, and muscle tissue, and you'd need something on the level of a mammogram to develop a robust picture of what would be necessary for proper support and fit. Not to be gross, but a source of naked breasts that does comes to mind is pornography. Feed that into the magic algorithm.
The algorithm would also struggle with "perfect", as that's too broad since different people want and need different things out of a bra. The founder, who is a woman, can spend some of the VC money to get herself a $1,000 bra (for "market research") but that's still only one data point, with her own preferences. (To be clear, I don't begrudge her that.) It seems like concentrating on a specific sub-set of the market to perfect the magic algorithm might cut down the problem to make it easier, though it would also cut down the size of the currently addressable market. There may not an existing huge market for $135 bras, but it seem targeting "women with breasts large enough to cause back problems". $135 doesn't seem like as much when compared to getting a two-hour massage from a licensed massage therapist or healthcare professional.
Alright though, let's lets hand-wave that they're able to perfect this magic algorithm, and are able to convince customers to send in the requisite data (aka as many photos of their breasts as it needed). How do you scale production of the bra's themselves, given the custom nature. Bra's are only as cheap as they are thanks to exploitation of workers in poorer countries, and garment manufacturing can't be automated with current technology, as discussed on the Blue a couple months back.
I must applaud the effort though, my non-existent company has't yet to sell any bras. Plus, it's at least a different idea than Uber, but for ________ , and it serves an under-served market. If men had breasts there would be a gajjillion options because men went out and made companies trying to address the need. We should be building up Mona Zhang, not tearing her down because she tried to tackle a hard problem, and failed.
posted by fragmede at 7:22 PM on December 25, 2019 [5 favorites]
Even if we hand-wave a magic algorithms could solve the problem, they just didn't have enough data. Nor do I see a way to get the level of data I think would be needed. "Please send us comprehensive photos of your breasts (including some with hands cradling them) . Additionally, we'll need video of them while you jump around, in order for us to create a perfect bra. We'll also need your name and shipping address." Uh huh. (Additionally, I don't believe pictures are sufficient enough, since the internal make-up varies between breast, fat, and muscle tissue, and you'd need something on the level of a mammogram to develop a robust picture of what would be necessary for proper support and fit. Not to be gross, but a source of naked breasts that does comes to mind is pornography. Feed that into the magic algorithm.
The algorithm would also struggle with "perfect", as that's too broad since different people want and need different things out of a bra. The founder, who is a woman, can spend some of the VC money to get herself a $1,000 bra (for "market research") but that's still only one data point, with her own preferences. (To be clear, I don't begrudge her that.) It seems like concentrating on a specific sub-set of the market to perfect the magic algorithm might cut down the problem to make it easier, though it would also cut down the size of the currently addressable market. There may not an existing huge market for $135 bras, but it seem targeting "women with breasts large enough to cause back problems". $135 doesn't seem like as much when compared to getting a two-hour massage from a licensed massage therapist or healthcare professional.
Alright though, let's lets hand-wave that they're able to perfect this magic algorithm, and are able to convince customers to send in the requisite data (aka as many photos of their breasts as it needed). How do you scale production of the bra's themselves, given the custom nature. Bra's are only as cheap as they are thanks to exploitation of workers in poorer countries, and garment manufacturing can't be automated with current technology, as discussed on the Blue a couple months back.
I must applaud the effort though, my non-existent company has't yet to sell any bras. Plus, it's at least a different idea than Uber, but for ________ , and it serves an under-served market. If men had breasts there would be a gajjillion options because men went out and made companies trying to address the need. We should be building up Mona Zhang, not tearing her down because she tried to tackle a hard problem, and failed.
posted by fragmede at 7:22 PM on December 25, 2019 [5 favorites]
Men do have breasts, but if more of them needed bras for fashion/social reasons, there would be three options - boob-briefs, boob-boxers, or boob-boxer-briefs.
posted by muddgirl at 7:32 PM on December 25, 2019 [5 favorites]
posted by muddgirl at 7:32 PM on December 25, 2019 [5 favorites]
I switched to a random sleep bra that looks like a sports bra made out of mostly Lycra and comes in S, M, X, XX, XXL with no cup sizes or bands. Just all elastic stretch. My breasts look vaguely squishy in it, so if I have a work thing where I need to appear culturally-socialized, I pull out my proper underwire bra and drag it on to shape my breasts into fashion-aesthetically appropriate appearance under clothes. Otherwise, I am now of the opinion fuck that. We have spandex and Lycra and elastic now and can be comfortable all the time even with size F breasts!
posted by dorothyisunderwood at 8:23 PM on December 25, 2019 [5 favorites]
posted by dorothyisunderwood at 8:23 PM on December 25, 2019 [5 favorites]
Also the bra cost $20.
posted by dorothyisunderwood at 8:24 PM on December 25, 2019 [1 favorite]
posted by dorothyisunderwood at 8:24 PM on December 25, 2019 [1 favorite]
The tailoring is a red herring here in a sense.
Is it, though? Tailoring seems the key to providing a bra suited to the wearer. Of course, nobody has to buy a bespoke bra, and probably most users don't.
Suit tailoring probably isn't as critical as bra fitting, a poorly fitting suit will suffice for many situations. Look around any business district at lunchtime, and you'll see quite a few men who simply do not wear their suits very well, where tailoring is minimal or sub optimal. But it just needs to feel well enough to the wearer and present well enough to the public. And even a less than tailored suit can still cost a pretty penny. The situation seem similar to the bra in that way.
The big difference isn't the value of tailoring, but that most men rarely have to wear a suit. I've worn one only a handful of times in my life, and it makes me sick when the situation arises with the need to get it altered. If I'm lucky, I'll be presentable when I wear it. The situation is rare enough that I don't care all that much if it doesn't hang terribly well on my frame. Most women, however, will have to wear a bra far more often than I'll ever have to wear a suit. Being an undergarment, comfort is at least as critical, if not more, than the way it presents.
If men had boobs, we'd have a billion options, all affordable, and a well understood and functional standard approach to bra sizing and fitting, as well as widespread and generally excellent tailoring services. Instead we get a half-assed system of just band and cup that kinda sorta maybeworks if you're near enough average, and god help you otherwise.
Is this true, though? The fine points of bra fitting seem much more complicated than the rather modular affair of suit fitting. In my profession, we'd say the tolerances are much more critical and the variables are vast and not very well understood, even by people in the industry who actually have boobs. In order for it to be viable, they have to get it right, all the time, with significant enough sales and profits. The barriers to a more automated and lower cost (and more profitable) approach to good bra manufacture appear to come from multiple fronts.
posted by 2N2222 at 9:13 PM on December 25, 2019 [1 favorite]
Is it, though? Tailoring seems the key to providing a bra suited to the wearer. Of course, nobody has to buy a bespoke bra, and probably most users don't.
Suit tailoring probably isn't as critical as bra fitting, a poorly fitting suit will suffice for many situations. Look around any business district at lunchtime, and you'll see quite a few men who simply do not wear their suits very well, where tailoring is minimal or sub optimal. But it just needs to feel well enough to the wearer and present well enough to the public. And even a less than tailored suit can still cost a pretty penny. The situation seem similar to the bra in that way.
The big difference isn't the value of tailoring, but that most men rarely have to wear a suit. I've worn one only a handful of times in my life, and it makes me sick when the situation arises with the need to get it altered. If I'm lucky, I'll be presentable when I wear it. The situation is rare enough that I don't care all that much if it doesn't hang terribly well on my frame. Most women, however, will have to wear a bra far more often than I'll ever have to wear a suit. Being an undergarment, comfort is at least as critical, if not more, than the way it presents.
If men had boobs, we'd have a billion options, all affordable, and a well understood and functional standard approach to bra sizing and fitting, as well as widespread and generally excellent tailoring services. Instead we get a half-assed system of just band and cup that kinda sorta maybeworks if you're near enough average, and god help you otherwise.
Is this true, though? The fine points of bra fitting seem much more complicated than the rather modular affair of suit fitting. In my profession, we'd say the tolerances are much more critical and the variables are vast and not very well understood, even by people in the industry who actually have boobs. In order for it to be viable, they have to get it right, all the time, with significant enough sales and profits. The barriers to a more automated and lower cost (and more profitable) approach to good bra manufacture appear to come from multiple fronts.
posted by 2N2222 at 9:13 PM on December 25, 2019 [1 favorite]
Not to be gross, but a source of naked breasts that does comes to mind is pornography. Feed that into the magic algorithm. What?!! I’m struck nearly wordless because the vast majority of pornography shows such a small variety of body types and so many of the breasts are augmented compared to non-porn actors which affects fit. Not to mention that 68% of American women wear above a size 14 and, while I just don’t feel like looking up the depressing real stats to compare, the number of porn actors who are that size or larger is far far less. Even the “BBW” in porn are often not that large! And how many of those actors are older than 30? 40? 50? How many have breastfed?
I know that was just one thought but it just horrified me. My bra choices are already dictated by a version of what my body is supposed to be shaped like or what it’s presumed I want from my breasts and pornography has an outsized influence already.
Daniel Lavery’s essay about wearing a binder made me want to try one really badly but I haven’t yet. Appreciate folks who mentioned them being potentially more comfortable than bras! Fshgrl, I’m also curious what bodice style bra you have and like if you’re up for sharing! I have loved some longline bras in the past but have had a hard time finding any without lace, which irritates my skin something awful in a way it didn’t used to when I was younger.
posted by the thorn bushes have roses at 9:46 PM on December 25, 2019 [1 favorite]
I know that was just one thought but it just horrified me. My bra choices are already dictated by a version of what my body is supposed to be shaped like or what it’s presumed I want from my breasts and pornography has an outsized influence already.
Daniel Lavery’s essay about wearing a binder made me want to try one really badly but I haven’t yet. Appreciate folks who mentioned them being potentially more comfortable than bras! Fshgrl, I’m also curious what bodice style bra you have and like if you’re up for sharing! I have loved some longline bras in the past but have had a hard time finding any without lace, which irritates my skin something awful in a way it didn’t used to when I was younger.
posted by the thorn bushes have roses at 9:46 PM on December 25, 2019 [1 favorite]
Are binders considered safe now? I ask as someone who would buy one for the sake of fashion versatility and settles for sports bras in the meantime.
posted by Selena777 at 10:15 PM on December 25, 2019
posted by Selena777 at 10:15 PM on December 25, 2019
Are binders considered safe now? I ask as someone who would buy one for the sake of fashion versatility and settles for sports bras in the meantime.
I'm talking a style of bra not an actual binder. Binder style sports bras have existed since lycra was invented, like the Cheeta brand, and before that people just wrapped their boobs. I think the safety factor is in how tight you wear those garments or bandages, not the idea of a flattening type chest garment itself per se. I'm not trying to make my boobs look invisible but what I mainly want from a sports type bra is to pin them down from the top and front evenly, which is where most sports bra fail spectacularly.
posted by fshgrl at 11:09 PM on December 25, 2019
I'm talking a style of bra not an actual binder. Binder style sports bras have existed since lycra was invented, like the Cheeta brand, and before that people just wrapped their boobs. I think the safety factor is in how tight you wear those garments or bandages, not the idea of a flattening type chest garment itself per se. I'm not trying to make my boobs look invisible but what I mainly want from a sports type bra is to pin them down from the top and front evenly, which is where most sports bra fail spectacularly.
posted by fshgrl at 11:09 PM on December 25, 2019
A friend of mine underwent a breast reduction surgery many years ago that her insurance paid for because she was in pain all the time. I’m glad she had that option. It feels wrong that so many people with breasts suffer from chronic discomfort because they don’t have access to appropriate bras. I understand this is a difficult problem to solve. It still sucks. Thanks for the post, OP.
posted by Bella Donna at 12:11 AM on December 26, 2019 [2 favorites]
posted by Bella Donna at 12:11 AM on December 26, 2019 [2 favorites]
but there has been plenty in the past and will continue to be going forward becuase women want and will pay for it and always have.
Fun fact: the first commercially available sports bra was made available in 1975! So maybe not quite "always".
posted by Gin and Broadband at 3:47 AM on December 26, 2019 [2 favorites]
Fun fact: the first commercially available sports bra was made available in 1975! So maybe not quite "always".
posted by Gin and Broadband at 3:47 AM on December 26, 2019 [2 favorites]
If men had boobs, we'd have a billion options
Sorry to be that guy but plenty of men have boobs, in particular trans men but also some cis men. I get what you mean but this felt crummy to read.
posted by an octopus IRL at 4:49 AM on December 26, 2019 [7 favorites]
Sorry to be that guy but plenty of men have boobs, in particular trans men but also some cis men. I get what you mean but this felt crummy to read.
posted by an octopus IRL at 4:49 AM on December 26, 2019 [7 favorites]
I really appreciate this writeup - and it makes total sense why I, as a post-breastfeeding large-back-large-breasts, am chronically unable to buy comfortable bras.
posted by corb at 6:42 AM on December 26, 2019 [1 favorite]
posted by corb at 6:42 AM on December 26, 2019 [1 favorite]
This is an obvious hard problem that maybe isn't solvable with current technology. Comparisons to suit tailoring are fundamentally flawed because suits just need to cover; they don't need to support. A suit would need to be over the top malfitting before it would cause discomfort via binding/chaffing/pinching/pressure. And a suit will still perform it's basic function of protecting the wearer from the elements and not letting things hang out even if so poorly fitted that it looks bad or if the wearer looses 10-50 pounds.
I'm skeptical that any visual/simple measurement system by itself will ever produce a decent custom bra for many of the reasons already laid out (differences in muscle mass/ breast tissue density/ attachment/ rib cage shape/ variability in end user expectations to name a few). The comparison to the equipment used by podiatrists to mould custom orthodics is apt.
I spent money on dozens of different off the shelf boot inserts over the years (and fancy and not so fancy athletic shoes) and all did little to nothing to mitigate foot and knee pain1. I finally acquired health insurance that let me get custom fitted inserts and the pain went away *poof*. But that involved seeing a trained specialist doctor; having my feet cast with a machine that measured pressures across the foot bed and then getting the inserts custom fitted afterwards. Total cost over $500 and I got ONE set of inserts. Additional inserts are still a couple hundred bucks and that is for a device that can be 80% moulded by machine rather than being essentially hand made. And the insert itself doesn't generally wear out though the top cushion layer requires occasional replacement.
And my feet essentially don't change over the course of a month or with relatively minor changes in body mass.
If a truly custom bra that will be well fitting and supportive for a wide variety of people is possible I bet it will end up being several hundred dollars and require more sophisticated equipment than a tape. And the average person will need several different versions to accommodate fluctuations.
An availability problem applies to many consumer goods with sveral variables even directed at men. Styles and markets change and unless something sells enough to make a healthy profit companies stop making them. Every once and a while a company puts out a product that by chance meets my need perfectly but doesn't meet enough people's needs and is therefor discontinued. My spouse and several of the commentators in this thread lament the perfect bra that has wore out and they can no longer buy. In a similar way the older I get the more I naturally acquire a list of perfect items I can no longer buy or have a greatly reduced range of choice. Small, cheap two seater 2DRHT coupes (there aren't any), small cheap two seater convertibles (there is the MX-5 and it's badge engineered cousin the Fiat 124; neither are particularly cheap but they are half the price or less of any of the other limited options); small pickups; button locking ratchet extensions (mine got stolen and I haven't been able to replace them); dozens of different hand planes; the old style cast marking craftsman sockets; 501s with the correct number of belt loops and buttons, etc. A buddies' shop burned down last year and it gave me nightmares for months. Leaving aside all the literally irreplaceable air dried wood he lost; the thought of the loss of the perfect tools that are either currently unobtainium or will require scouring ebay and other auction sites for used versions is disheartening. Many craftsmen have tools or at least jigs they have made themselves because no one makes or has ever made a tool to do a particular job in a particular way. The fact that their isn't a cottage industry or even hobbyist network of people making their own well fitting, supportive bras goes to show just how fundamentally hard the problem is.
PPS: "Not to mention that 68% of American women wear above a size 14 and, while I just don’t feel like looking up the depressing real stats to compare, the number of porn actors who are that size or larger is far far less. Even the “BBW” in porn are often not that large! And how many of those actors are older than 30? 40? 50? How many have breastfed?"
While problematic for many reasons lack of source material is not one of them. There is a lot of porn/erotica available and one of the areas that can turn serious profit is niche markets. While the majority features currently conventionally attractive actresses once you start looking any body type and age range is represented. _Any_. Old women (both porn old and actual old) easy. Unaugmented or breastfed? So common as to not even be notable; currently lactating is also widely available. Size 14, 24, 34 also common Rule 34 is observational not dictatorial. A random sampling wouldn't get a spectrum but a directed effort would. If a person wanted to acquire naked photos of a wide range of bodies from various angles porn would be a viable way to do so with a little work.
[1] and even if I had managed to find the perfect shoe by fluke it wouldn't be available next year. When I was in my early teens I once spent an entire weekend with an uncle driving around to different shoes stores buying the exact same mass produced shoe in his size (way before the internet). I think he ended up buying like 47 pairs. It wasn't until I was in my twenties that I really grokked the value of well fitting shoe (or anything you can't make yourself or afford to get custom made really) and why my uncle was buying a life time supply of a single model of athletic shoe.
posted by Mitheral at 10:29 AM on December 26, 2019 [3 favorites]
I'm skeptical that any visual/simple measurement system by itself will ever produce a decent custom bra for many of the reasons already laid out (differences in muscle mass/ breast tissue density/ attachment/ rib cage shape/ variability in end user expectations to name a few). The comparison to the equipment used by podiatrists to mould custom orthodics is apt.
I spent money on dozens of different off the shelf boot inserts over the years (and fancy and not so fancy athletic shoes) and all did little to nothing to mitigate foot and knee pain1. I finally acquired health insurance that let me get custom fitted inserts and the pain went away *poof*. But that involved seeing a trained specialist doctor; having my feet cast with a machine that measured pressures across the foot bed and then getting the inserts custom fitted afterwards. Total cost over $500 and I got ONE set of inserts. Additional inserts are still a couple hundred bucks and that is for a device that can be 80% moulded by machine rather than being essentially hand made. And the insert itself doesn't generally wear out though the top cushion layer requires occasional replacement.
And my feet essentially don't change over the course of a month or with relatively minor changes in body mass.
If a truly custom bra that will be well fitting and supportive for a wide variety of people is possible I bet it will end up being several hundred dollars and require more sophisticated equipment than a tape. And the average person will need several different versions to accommodate fluctuations.
An availability problem applies to many consumer goods with sveral variables even directed at men. Styles and markets change and unless something sells enough to make a healthy profit companies stop making them. Every once and a while a company puts out a product that by chance meets my need perfectly but doesn't meet enough people's needs and is therefor discontinued. My spouse and several of the commentators in this thread lament the perfect bra that has wore out and they can no longer buy. In a similar way the older I get the more I naturally acquire a list of perfect items I can no longer buy or have a greatly reduced range of choice. Small, cheap two seater 2DRHT coupes (there aren't any), small cheap two seater convertibles (there is the MX-5 and it's badge engineered cousin the Fiat 124; neither are particularly cheap but they are half the price or less of any of the other limited options); small pickups; button locking ratchet extensions (mine got stolen and I haven't been able to replace them); dozens of different hand planes; the old style cast marking craftsman sockets; 501s with the correct number of belt loops and buttons, etc. A buddies' shop burned down last year and it gave me nightmares for months. Leaving aside all the literally irreplaceable air dried wood he lost; the thought of the loss of the perfect tools that are either currently unobtainium or will require scouring ebay and other auction sites for used versions is disheartening. Many craftsmen have tools or at least jigs they have made themselves because no one makes or has ever made a tool to do a particular job in a particular way. The fact that their isn't a cottage industry or even hobbyist network of people making their own well fitting, supportive bras goes to show just how fundamentally hard the problem is.
PPS: "Not to mention that 68% of American women wear above a size 14 and, while I just don’t feel like looking up the depressing real stats to compare, the number of porn actors who are that size or larger is far far less. Even the “BBW” in porn are often not that large! And how many of those actors are older than 30? 40? 50? How many have breastfed?"
While problematic for many reasons lack of source material is not one of them. There is a lot of porn/erotica available and one of the areas that can turn serious profit is niche markets. While the majority features currently conventionally attractive actresses once you start looking any body type and age range is represented. _Any_. Old women (both porn old and actual old) easy. Unaugmented or breastfed? So common as to not even be notable; currently lactating is also widely available. Size 14, 24, 34 also common Rule 34 is observational not dictatorial. A random sampling wouldn't get a spectrum but a directed effort would. If a person wanted to acquire naked photos of a wide range of bodies from various angles porn would be a viable way to do so with a little work.
[1] and even if I had managed to find the perfect shoe by fluke it wouldn't be available next year. When I was in my early teens I once spent an entire weekend with an uncle driving around to different shoes stores buying the exact same mass produced shoe in his size (way before the internet). I think he ended up buying like 47 pairs. It wasn't until I was in my twenties that I really grokked the value of well fitting shoe (or anything you can't make yourself or afford to get custom made really) and why my uncle was buying a life time supply of a single model of athletic shoe.
posted by Mitheral at 10:29 AM on December 26, 2019 [3 favorites]
So maybe the answer (eventually) is smart materials which can adapt rather than making a basically static bra.
posted by Nancy Lebovitz at 10:32 AM on December 26, 2019 [2 favorites]
posted by Nancy Lebovitz at 10:32 AM on December 26, 2019 [2 favorites]
Are binders considered safe now?I was going to say yes, and then I remembered this study from 2017 that found that 97% of people who bind regularly have had at least one negative side effect from binding. Still, if you use a binder that isn't too small, stay away from duct tape or ace bandages, and don't bind 24/7, the risks are fairly low. I like gc2b's design, which includes a panel of heavy, non-elastic fabric in the front that does a lot of the work of flattening without the binder itself needing to be super tight.
posted by Basil Stag Hare at 10:41 AM on December 26, 2019
Part of the issue is preferences. Fshgrl hates a tight bra, I can't stand one that is loose!
I am curious about the Bobbie and Evelyn bra mentioned above, but looking at it, I can't imagine anything sized xs-l would work for me.
posted by vespabelle at 10:44 AM on December 26, 2019
I am curious about the Bobbie and Evelyn bra mentioned above, but looking at it, I can't imagine anything sized xs-l would work for me.
posted by vespabelle at 10:44 AM on December 26, 2019
Information about sizing a bra properly is very much available in the year 2019 and if you want to start a bra company without learning about that maybe... I don’t know.
"Sizing a bra" is basically determining which off-the-shelf bra size is going to be least disruptive to one's actual figure. It generally involves taking measurements and then trying on many bras which purportedly "fit" those measurements and then seeing which might actually fit. It's not the least bit scientific, and you can't build an algo from that sort of information.
This company was not trying to solve that problem; they were trying to build a system that would take numerous data points and create a custom product. That is not "sizing a bra". As many people pointed out, comparing this process to bespoke, custom, or semi-custon suits is not the same. A suit is a shell; a bra is meant to be like a supportive second skin. You have to deal not just with circumference but with volume, and where that volume is located. You have to deal with physics. Furthermore, bras are a twentieth century invention, and in the last 1000 years what people expect from bras and the people that wear them has changed radically. No one expects a person wearing a suit to dig a ditch or run a 10k race, but we expect people wearing bras to do those things and and those people would like to remain comfortable and feel supported, sometimes discrete and always pain-free whether they wear a bra for a few hours a day or all day, seven days a week. Who has those expectations of a suit?
posted by oneirodynia at 10:58 AM on December 26, 2019 [2 favorites]
"Sizing a bra" is basically determining which off-the-shelf bra size is going to be least disruptive to one's actual figure. It generally involves taking measurements and then trying on many bras which purportedly "fit" those measurements and then seeing which might actually fit. It's not the least bit scientific, and you can't build an algo from that sort of information.
This company was not trying to solve that problem; they were trying to build a system that would take numerous data points and create a custom product. That is not "sizing a bra". As many people pointed out, comparing this process to bespoke, custom, or semi-custon suits is not the same. A suit is a shell; a bra is meant to be like a supportive second skin. You have to deal not just with circumference but with volume, and where that volume is located. You have to deal with physics. Furthermore, bras are a twentieth century invention, and in the last 1000 years what people expect from bras and the people that wear them has changed radically. No one expects a person wearing a suit to dig a ditch or run a 10k race, but we expect people wearing bras to do those things and and those people would like to remain comfortable and feel supported, sometimes discrete and always pain-free whether they wear a bra for a few hours a day or all day, seven days a week. Who has those expectations of a suit?
posted by oneirodynia at 10:58 AM on December 26, 2019 [2 favorites]
Sorry to be that guy but plenty of men have boobs, in particular trans men but also some cis men. I get what you mean but this felt crummy to read.
Please don't apologise, you're completely correct to call me out on that, and I'm very sorry for fucking that up so badly.
posted by Dysk at 1:11 PM on December 26, 2019
Please don't apologise, you're completely correct to call me out on that, and I'm very sorry for fucking that up so badly.
posted by Dysk at 1:11 PM on December 26, 2019
If the problem is too hard to solve, then it won't be seen as a problem at all for those with bodies privileged by patriarchy. There's an emperor's new clothes effect where roughly half the population can let their nipples show through their polo shirts and go swimming bare-chested, and roughly half would cause immense scandal if they tried.
posted by Former Congressional Representative Lenny Lemming at 2:20 PM on December 26, 2019 [6 favorites]
posted by Former Congressional Representative Lenny Lemming at 2:20 PM on December 26, 2019 [6 favorites]
vespabelle, is this the bra you’re thinking might not work for you? It’s the one I was talking about and I don’t know if you mean it might be too big or too small, but I wear a 3-4x in most clothes and am a 38J cup and my friend is a 1x and I think a 42C cup so the one she gave me is a 1x and it works for me for some reason. Don’t know if that helps since essentially their size chart doesn’t really seem accurate. (My friend, the fit model, thought it put too much pressure on her shoulders — which I don’t feel at all — but otherwise found the support and fit of the cups was great and so did I, but it was hard to get on at first.)
Also the gc2b is the brand that Daniel Lavery waxed poetic about that made me think of trying one. I can’t find the essay but I think it was something I first saw on Metafilter! I’m on the femme side of things and minimizing bras were always awful so a binder would never have occurred to me. I’m not the target market but as with universal design a well made solution can benefit lots of folks.
posted by the thorn bushes have roses at 3:30 PM on December 26, 2019
Also the gc2b is the brand that Daniel Lavery waxed poetic about that made me think of trying one. I can’t find the essay but I think it was something I first saw on Metafilter! I’m on the femme side of things and minimizing bras were always awful so a binder would never have occurred to me. I’m not the target market but as with universal design a well made solution can benefit lots of folks.
posted by the thorn bushes have roses at 3:30 PM on December 26, 2019
This might be the Lavery article. It’s under a prior surname.
posted by Selena777 at 4:47 PM on December 26, 2019
posted by Selena777 at 4:47 PM on December 26, 2019
Please don't apologise, you're completely correct to call me out on that, and I'm very sorry for fucking that up so badly.
Thanks (:
posted by an octopus IRL at 5:07 PM on December 26, 2019
Thanks (:
posted by an octopus IRL at 5:07 PM on December 26, 2019
That's it, Selena777, thank you! My search skills on mobile are weak even though I knew the name it was published under was an older one!
posted by the thorn bushes have roses at 5:28 PM on December 26, 2019
posted by the thorn bushes have roses at 5:28 PM on December 26, 2019
Excellent, carefully fitted bras are being made in Poland.
posted by Nancy Lebovitz at 8:38 PM on December 26, 2019
posted by Nancy Lebovitz at 8:38 PM on December 26, 2019
Part of the issue is preferences. Fshgrl hates a tight bra, I can't stand one that is loose!
Oh I don't like it loose either but the Bra That Fits guidelines put many people into bras that are waaaaayyy too tight.
posted by fshgrl at 10:18 PM on December 26, 2019 [1 favorite]
Oh I don't like it loose either but the Bra That Fits guidelines put many people into bras that are waaaaayyy too tight.
posted by fshgrl at 10:18 PM on December 26, 2019 [1 favorite]
If the problem is too hard to solve, then it won't be seen as a problem at all for those with bodies privileged by patriarchy.
I’ve been wondering how this would play out, and am imagining that people would buy whatever underwear was most comfortable and then invest in suits that draped elegantly over the result. After all, this is what the spheroidal tycoon on the Monopoly box has done. But also it’s how the classes of women who could age into power dress or used to dress - `clubwomen' 1880-1950, Black women who were powerful in church or public service in my childhood. It’s kind of the market Eileen Fisher is aiming for, but her drapy materials are more revealing than a suit although easier to mass manufacture.
posted by clew at 1:26 PM on December 28, 2019 [1 favorite]
I’ve been wondering how this would play out, and am imagining that people would buy whatever underwear was most comfortable and then invest in suits that draped elegantly over the result. After all, this is what the spheroidal tycoon on the Monopoly box has done. But also it’s how the classes of women who could age into power dress or used to dress - `clubwomen' 1880-1950, Black women who were powerful in church or public service in my childhood. It’s kind of the market Eileen Fisher is aiming for, but her drapy materials are more revealing than a suit although easier to mass manufacture.
posted by clew at 1:26 PM on December 28, 2019 [1 favorite]
the thorn bushes have roses: ... in order to get a lot of support out of the (bra) band I have to put up with deep welts. According to A Bra That Fits that is normal and okay ...
I belong to A Bra That Fits, and holy hell, "deep welts are an acceptable trade-off for a supportive bra" is not the message I'm sharing with people who ask for assistance. It's "A Bra That Fits," not "A Bra That Hurts." I'm sorry you went through all that.
posted by virago at 7:23 PM on December 30, 2019
I belong to A Bra That Fits, and holy hell, "deep welts are an acceptable trade-off for a supportive bra" is not the message I'm sharing with people who ask for assistance. It's "A Bra That Fits," not "A Bra That Hurts." I'm sorry you went through all that.
posted by virago at 7:23 PM on December 30, 2019
there is a reason lace up supporters were so popular for so long
This might be the big thing I've taken from this thread: bras as currently designed are probably the wrong solution for a lot of people. A better-fitting version of the wrong thing won't help. Binders and corsets and I dunno, stretchy boob-socks or something aren't fashionable but might do a better job of meeting the wearer's needs. Maybe we need more variety based on purpose (support, concealment, comfort) instead of fashion.
posted by harriet vane at 5:47 AM on December 31, 2019 [2 favorites]
This might be the big thing I've taken from this thread: bras as currently designed are probably the wrong solution for a lot of people. A better-fitting version of the wrong thing won't help. Binders and corsets and I dunno, stretchy boob-socks or something aren't fashionable but might do a better job of meeting the wearer's needs. Maybe we need more variety based on purpose (support, concealment, comfort) instead of fashion.
posted by harriet vane at 5:47 AM on December 31, 2019 [2 favorites]
« Older 1939 MGM cartoon by Hugh Harman | Distant Sky Live Newer »
This thread has been archived and is closed to new comments
posted by GenjiandProust at 6:19 AM on December 25, 2019