Hacker News Reader

Comment by Maciek416 | Hacker News Reader

This could present a significant barrier to reliable online shopping for clothes. It'll be interesting to see in the coming years, as more of these retailers offer online ordering (GAP is finally offering their catalog to us Canadians, for example), whether fitting will be normalized somehow, or whether each retailer will instead stick to a specific size translation table as I've seen some places.

Either that or someone will have to build a website or tool that aggregates the investigative work for us. I know I'd pay a small premium to eliminate the guesswork.

(EDIT: I wish there was a way to OpenID-ify personal body measurements for this purpose so every clothing site I visited could automatically select the best-fitting clothes for me. GAP small/medium t-shirts are wildly different from Threadless' small/medium American Apparel Ts)

View on HN

> build a website or tool that aggregates the investigative work for us

My bane has always been t-shirts. I can take anywhere from a medium to x-large depending on the brand. After the first wash, the shirt will change size drastically or hardly at all. Buying t-shirts is a complete $20 gamble, I have no idea if ultimately the shirt will fit me or not. At least pants, dress shirts and shoes you can be reasonably confident they will fit you when you try them on at the store.

I've always yearned for some kind of solution to this, but admittedly have never really came up with a good one.

Same situation for me. I purchased some XL tshirts at Old Navy a year back, and I can't wear them. It's one thing if an XL fits more like a L, but these are like a M.

It's hard enough to find well-fitting clothing when you can try it on. I'm not going to be ordering over the net anytime soon.

My friends' startup is tackling this issue for shoes, using lasers: http://shoefitr.com/

The interface is really, really well done, check out their sample page: http://shoefitr.com/mock_checkout.php and click 'let Shoefitr help you choose the right size.'

Interesting... this would be really helpful if it also took into account pronation (see http://www.runnersworld.com/article/0,7120,s6-240-319-327-77...).

I will send that to them, thanks.

The only problem is, that shoes that try to target different pronation types actually cause more injuries.

(http://www.sportsci.org/jour/0103/mw.htm#_Toc535425250 cites some studies beginning with "Wearers of expensive running shoes that were promoted as correcting pronation or providing more cushioning experienced a greater prevalence of these running-related injuries than wearers of less expensive shoes [...]")

I've been waiting for something like this for shoes. Looks excellent, though I'd recommend against the shrinking between operations when finding a size (primarily in the first step). Just takes longer and looks clunky.

I'll pass the feedback along.

Oh, and if you actually want to try it on a real site with real shoes, RunningWarehouse.com has them up, like this, for example:

http://www.runningwarehouse.com/descpageMRS-NVM5SBG.html

That's an amazing interface but it needs more brands and models in there, everything I tried came up 'sorry, we don't have that model scanned' or something to that effect.

But it's one of the nicer 'live' (ajax) interfaces that I've seen.

Out of curiosity, were you inputting running shoes? I think they've decided to focus on just running shoes first, and while they've been hard at it, it physically takes time and effort to get the raw data.

Yep.

> it physically takes time and effort to get the raw data.

I can see that... btw, the Reebok brand is one of the more popular ones and completely missing.

Again, kudos on the interface to your friend, I love it when people use new tech in non-fluffy ways (as in useful rather than eye candy). (and it looks great too!)

Many online clothing stores offer a liberal return policy to allow you to try them on.

If I have to mail it back, I've already wasted more time than I would have just going to a specialty store within 45 minutes of me.

Soon I'll be more like 4 hours from the nearest retailer of any sort, so that may change. But I'm somewhat unique in that regard.

Wow. I had a long, detailed conversation last night with my roommate/potential co-founder about doing EXACTLY this. The way sizing is handled in the clothing industry is ridiculous for women especially, but for men as well according to this article.

The basic technology product is software that, given information about your body shape, can produce a listing of clothing that will fit you. As far as we can tell this does not yet exist. When the customer can trust that the clothes will fit, then they can shop based on style, fashion, price, or any other number of more interesting criteria rather than worrying about getting the right size or cut.

If you pair this technology with a customer-centered online retail environment then I think you can create a lot of value. Zappos has shown that you can make buying articles of clothing online work, with their famous customer support and return policies.

The concept of requiring a customer to consult a sizing translation table per label or even per garment just to find clothes that fit well flabbergasts me. But we haven't found any better solutions on offer.

This all fits in with Request For Startup 2: New Paths Through Product Space. http://ycombinator.com/rfs2.html Nobody is selling clothing based on what fits the individual shopper. We want to make an online storefront that only offers clothing that will fit you.

There are a lot of negative knock-on effects from the current way clothes shopping is done. People agonize over sizing and equate their waistband with their self-worth. That's ridiculous. Every person will look best and be most comfortable in clothing that matches their actual shape, regardless of what number is on the label. Especially if that number is a lie!

Clothes shopping changed drastically starting in about the mid 1800s through the 1920s with the advent of mass-produced garments. Before that each item was made individually. After that, confusing labels, garments designed to fit the "average" person to minimize the need for alterations, changing demographics, and vanity sizing turned the whole system into an incredible morass. It wasn't always this way, and it shouldn't have to be.

The article suggests that government regulation of sizing labels is the solution to the problem of ill-fitting clothing. What the author does not mention is that this was tried in the past! That system fell apart and you can read about its failure from the NIST: http://museum.nist.gov/exhibits/apparel/index.htm

I posit that smarter technology which can match the clothing to the person is a superior solution.

There's a lot of pain involved in clothes shopping today. We want to fix that.

If any hackers here, especially women, have some thoughts to share on this topic I dearly want to hear them, whether in a post here or in a private email. My address is in my profile.

We are seriously considering applying to YCombinator with this proposal.

I have a few observations that might be worth considering.

1. For most women clothes shopping is a recreational activity, not a chore. While most men would be happy to minimize their shopping time, most women would be happy to extend it. Likewise, on the seller side, stores maximize women's departments for impulse buying, not efficiency. It's not rational or productive, but that's the way it is. I'd be wary of a software solution that solves a problem that both sellers and buyers should have but don't.

2. If you've ever made clothing patterns or have been custom fitted by a professionally-trained tailor, you would know that it takes about 3 fittings to get a piece right. This is after your personal measurements have been taken on the initial visit. To me, that signals "human problem" (a fuzzy intelligence task) rather than "computer problem" (a highly repetitive task with clearly defined parameters).

3. There are hundreds of fit variations, which combined with size variations and body types, produce a million different results. Natural-waist pants have a different waist size than low-waist pants. A jacket with a shrunken fit has higher armholes so the arm circumference and arm length would have to be adjusted. A woman with a long torso would probably pick natural-waist jeans and a shrunken-fit jacket if she wanted the low-waist jeans with a bolero jacket look. And so on, ad infinitum.

4. Manufacturers do not make different clothes to sell online vs. in-store, but rather they make clothes to sell in-store and then sell those online. The reality of in-store clothes is, it is made to look good on a hanger, not on a woman. Which is why models are as humanly close to a hanger as possible. The math is, a woman will grab 10 items that only look good on a hanger/model, and maybe buy 2 items after trying them on; conversely, a woman will grab 0 items that only look good on a real body, and buy 0 items as a result.

On a separate note, my husband works in the fashion industry, and a big part of his job is collaborating on the design and manufacturing process as far as fit etc. I'd be happy to pass on any questions you might have.

in re: 1

My wife is a doctor who needs to buy a fair amount of clothes and, I assure you, does not find shopping to be a recreational activity.

Also, may not online shopping be recreational activity?

I add these points merely to suggest that there may be enough market in 1) women who don't like shopping and 2) women who recreationally shop online.

I get your point about how busy your wife is, but do you not see the irony? Your doctor wife "needs" to buy a lot of clothes? What, to put under her white lab coat where no one will see them?

Nobody "needs" a lot of clothes. The fact that both you and she think she does is just an artifact of internalizing the recreational drug that is acquisition of shiny things. So your wife's movie is that she that she needs to live up to the public image of a well-respected doctor, and my movie used to be that as a female techie, I need conservative, well-made things to tone down the sex appeal in order to be taken seriously at work. Name any profession, it's not hard to come up with plausible reasons to shop whereas in reality, apart from jobs in fashion, there is little correlation between career and clothes... and even there, you have your models' uniform of jeans with a white tanktop and everyone else in black.

As far as online shopping, yes, it's a recreational activity well-known in the e-commerce world. Shopping sites like Saks and Neiman Marcus get huge traffic spikes on weekend nights, i.e. the time of recreation (just google "drunk online shopping"). However they are in the same boat as the physical stores. If they start optimizing for efficiency, they will lose the impulse buyer, a huge chunk of their sales.

All in all, my point was not that 100% of women are recreational shoppers, even though we overwhelmingly are. It was to bring to the OP's attention the reality of their market. I've wasted my time on a startup that was a time sink similar in the sense that it optimized a big ugly inefficiency that our market just didn't want optimized and my post was motivated by empathy rather than negativity. I still don't want to say categorically that there isn't an opportunity here but I think it's important to go in with a clear head so I wanted to put in my 2 cents, especailly since the OP asked for some female perspective.

I don't think it's fair of you to accuse us of being shiny thing gatherers; I think we're far from that.

That said, working in the hospital means often getting blood/spit/worse on your clothes and needing to buy more. It also means having to dress nicely due to being a professional. She'd rather always wear scrubs, but she can't, and the white coat doesn't always protect her from everything.

My point was that there may be more market than you're suggesting there is.

I didn't argue with you about the recreational thing, but don't you think it's possible to make a site like the OP suggested without making shopping unfun or not recreational? It seems to me you could actually enhance the appeal of recreational shopping if you really thought you could trust the fit of items you were purchasing.

That's so strange! I've never seen a doctor without scrubs, either in an office or a hospital. Isn't blood etc. the reason doctors wear them in the first place? Are you in the US?

My intent was not to accuse you of being shiny things gatherers, sorry if it was unclear. I was speaking in the context of the discussion which was, is fit software a good idea, and you countered with, my doctor wife buys lots of clothes so she'd be a customer (if I did understand your point). If she just needs throw-away clothes to be soiled on a daily basis, she for sure doesn't need software to make them fit perfectly, am I wrong?

> I've never seen a doctor without scrubs, either in an office or a hospital.

Weird... your primary care physician wears scrubs? Pretty much only ER and Trauma docs wear scrubs here in Baltimore. (It was the same in New Haven).

Right now, she basically solves the clothing size problem with retailer loyalty: she knows what size Ann Taylor clothes fit her correctly. Perhaps if she was confident that she could figure out the correct size of items on the internet she'd feel comfortable shopping there.

(sorry to duplicate my comment here, but it's relevant)

It's not just a matter of vanity, it's a matter of extremely loose tolerances in manufacturing.

I've found that even within a single store, looking at the same model of Levi's jeans, I need to try each individual pair. I always wear the same thing: Levi's 505 34W 30L. But I find that individual specimens with this exact tag can be as much as an inch different in every dimension: waist diameter, inseam length, and even leg diameter at the hem.

This being the case, it's not possible to say that a given item is the right fit, without actually trying it.

It's not loose tolerances in manufacturing--it's just shitty quality control.

Sorry to be so vulgar, but it's true. Tons of boutiques and higher-end clothing labels have almost ZERO variance in sizing. Some of them hand-make their clothing, sure, but most don't.

Regardless, though, you're right--there's too much variance in sizing for mainstream brands. This problem really hinders any type of "OpenID-for-clothing" startup.

But if you have a clothing portal that knows your every dimension, part of the service you can sell is that you measure every single item, ensuring a good fit (regardless of the size tag). Whether you match them up to the customer on shipping or to the database on receiving is up to you.

I have no idea if that would be scalable, but it would certainly help with customer retention and goodwill.

You could always pitch it to a wealthier crowd to limit the amount of customers...

It can’t be that expensive (relative to making the clothing) to measure each article precisely and then bin them by actual size.

I've also found that Levi's have become so bad in this regard that I had to change to Lee's which seem much better. I think Levi's quality control went away about when they started selling to Walmart.

This is valuable information. Do you have any insight into the manufacturing process for garments? I don't yet know enough about the clothing industry and manufacturing so any data points are very welcome. I firmly believe that problems like these must be surmountable. We have the technology to build devices with tolerances measured in microns. Surely we can find a way to get clothes to fit.

It simply isn't practical. We have technology to build devices with micron tolerances, but they cost millions of dollars. Sure, you could build a device to sew a pair of jeans with millimeter tolerances, but you'd have to be so precise that you'd have to charge thousands for a single pair. That's not even looking at the material itself. Different fabrics can have different tolerances for stretching in all directions.

It is a big tradeoff between accuracy and cost. If you want to have the measurements be hyper-accurate, be prepared to spend a ton.

Large brands like Levi's generally use several contractors, and the quality (and therefore fit tolerance) inevitably varies from contractor to contractor. Chances are that was a bad batch and/or you were shopping in an outlet mall (where they send a lot of the odds and sods).

Perhaps, they should find a way to (re)measure the final product, and re-label as necessary.

Here's all I want:

Allow me to input something I know, e.g. I wear size 12 Adidas. Then tell me what size I need to buy of the shoes you're selling me, be they Nike, Reebok, or something new.

Shoes is one example, applies for many other types of clothing.

Euro sizing has been pretty reliable in my experience (except with gloves, but my hands are very slender and my knuckles rather broad). You could try translating the US size to Euro sizing based on Adidas' chart, and then converting Euro sizing to Nike using Nike's chart.

I think it's overall a great idea. Especially if the retailer can go beyond the basics and supply reliable information about the fit/cut of garments, since that is such an important element of style.

Some retailers already do this as a matter of course on their labeling in-store: Tags on H&M pants, for instance, are much better than average at providing detailed fit/cut info.

A potential sticking point that I can imagine is that some people will be reluctant to know their true measurements, and others might be wary of that information being stored online with their ID. (Imagine a famous actress struggling with her weight, only to have her hacked measurements exposed on some gossip blog.) Consider also, that many women are not going to want others to know their true measurements just from a glance at their laundry, so labels inside the garment should probably avoid using inches or centimeters, just as they do now for women's sizing in the US.

If any of that is hard for you to imagine, check to see if you are male. These are real issues. Females contend with a vast number of societal pressures concerning body image and beauty that men rarely need to consider for themselves.

This product does exist. See: myshape.com

If anyone wants to discuss this product in depth, drop me a line. This is a product space I have been eying for some time and have done considerable research and experimentation on the topic.

The basic technology product is software that, given information about your body shape, can produce a listing of clothing that will fit you. As far as we can tell this does not yet exist. When the customer can trust that the clothes will fit, then they can shop based on style, fashion, price, or any other number of more interesting criteria rather than worrying about getting the right size or cut.

I am too tall for an average clothes shop, so is my dad, and we were discussing something similar recently: our experience of clothes shopping is very similar to what you describe - walk into a shop, ask or search primarily "what do they have which might fit".

If you could do "only show me things which fit" from multiple shops in one place, that would be great, but fit is more than just leg length, waist size and brand, and a) I don't know what else matters so I couldn't give you said information about me, and b) even recently when I went to a shop and found trousers to fit, they did technically fit but they weren't comfortable and I wouldn't wear them.

Are you going to be able to do "things which will fit" instead of "things which might fit"? I find the latter is often barely different from "things which wont fit".

Incidentally, http://www.leftfootcompany.com/ has shops around where you get your feet scanned, then you leave and order on the internet, and they mechanically manufacture shoes to fit. Although they couldn't make shoes to fit me at the time I tried, which was unhelpful.

Most clothing sites have a sizing table, that lists the hip, waist, inseam, neck, etc for each of their sizes. The hips for women's large seems to vary from 30-42, but if the sizing chart matches my measurements, the clothes usually fit.

For women it's also confusing because it's often not clear if something is in juniors, misses, or women's sizes. But the size chart usually makes it clear.

If you build it, I'll use it.

One thing I found interesting in this vein was that there are only a handful of measurements that specify a pair of glasses (like pupil distance) and once you know yours, you can just buy glasses for $8, but very very few people know they can just ask for them so they buy $200 glasses from their optometrist.

In my experience, it's already a problem. When I'm shopping online with a brand whose fit I don't know, I order everything in medium and large and then return the size that fits less well. I'd imagine most online clothing stores have sky-high return rates.

I don't think many brands would want to do this, but it would help if they had a chart showing how their fit compares to other brands. E.g. "much looser than Dolce & Gabbana, looser than Calvin Klein, tighter than Gap".

I've basically given up buying clothing from the US (living in Europe). T-shirts in particularly are enormous by European standards.

I have neither a tiny nor a particularly big frame, but I feel tiny when I get the shirt I ordered online.

I've never had a problem with it. I wear 31x34. Then again, I only ever buy Levi's, so obviously I don't experience any brand variance. I'd be interested to see where they stack up on the chart.

mec.ca online store has something like this.

You rate the fit on several sliding scales (too long/too short etc) then buyers can estimate = I should buy brand X a size larger/smaller.

Reply on news.ycombinator.com