99Dresses (YC W12) wants to give women an infinite closet
techcrunch.com
techcrunch.com
By introducing virtual currency, the money put into the 99dresses stays in the 99dresses system. If real money were traded, a person could easily dump all of their unwanted clothes and be done with it. This isn't what 99dresses want. They want to build a community. They don't want to lose you after dumping your clothes - they want you to engage and interact by picking up secondhand clothes from other members of the community. They want to be your primary clothing provider - or close to it.
It's also worth noting that spending 'buttons' doesn't feel like spending money (even though they are equivalent to real money), which leads to impulse decisions and less friction. For the users, it feels more like an exchange than pouring in your hard earned money. The obstacle, of course, is drawing in new users, but once they're in, they're in.
"Dear Mark, Why is it that casinos use chips versus real money on most of their games? Dave M.
Because most players do not equate casino chips with real money. It is far too easy, Dave, to be caught up in the game and forget what you are actually betting. Chips possess none of the qualities of real money. A dollar has a more palpable feeling and is more difficult to surrender than parting with a casino chip. Real dollars evoke those hours of employment, mortgages, and dry cleaning bills. But once purchased, chips seem like fun tokens found at the fair, not a medium of exchange. That is why casinos prefer pit employees to "change color" or upgrade your chips. They are not being courteous, just trying to induce larger play."
source(down):http://www.markpilarski.com/column14-2.html
If the virtual currency is worth real money, then the company issuing the virtual currency can make money by letting people buy into the system. I'd expect 99Dresses to start allowing people to buy "buttons" with real money after that have built up enough inventory in the system.
(Outside the edit window, and now I've actually read the TechCrunch piece properly)
Why not allow people to exchange buttons back for money, just at a lower rate?
Why? When you receive a dress, it is set to a disabled state on the site (rendering it unavailable for others). I'm guessing a dress can be in this state indefinitely, if one desires to keep a dress forever.
> I might not even get to choose a dress I want if I'm worried that I won't be able to trade it again
When would this happen? Why would you not be able to trade it again?
It sounds like you're confused about the logistics in play. The system is effectively the same as any other buying and selling marketplace, but virtual currency is used in place of dollars.
You're right - you can keep a dress forever and it's not on the website. That's not really what I mean by "out of the system". Dresses need to be able to come out of the system because they go out of style, or get too used, and nobody wants them again. If I'm trading dresses and then get stuck with one, in a way it's still "in the system" and clogging it up. I can't participate until I buy back in again.
On the other hand, I already can't really remember why I blamed the virtual currency for this particular problem, so you're right about that, too. This problem exists when you use dollars, too.
You upload a dress and set your own price for it in buttons. When someone requests that dress you send it to them and you gain n buttons.
This would be ideal for her - however I think the button thing where you get button credits, but when it says "the seller pays for shipping and handling" this seems a little like a disincentive.
Do you get button credits for the shipping costs?
Who determines the button value of a given dress? The site or the seller?
Whats the typical shipping cost - and are savvy selleres accounting for this in their set button price?
It would be a hell of a lot smoother if you would print out a shipping label from 99dresses which has the buyers address etc on it. The cost comes out of the buyers button credits etc..
When I buy something from Amazon, I pay the shipping cost as the buyer - why the heck is it good to make the seller pay this, get a box fill out the label etc...
This alone will prevent my wife from using this service. Even though she is an ideal candidate.
I agree 100%, this is the killer app. My wife independently suggested this idea after buying some clothes online that technically were the right size but didnt provide a good fit for her body. I told her that it was a very difficult problem from a computer science perspective. Have you got any links (to Siggraph or others) which have taken a look at this problem in detail?
Aside from the obvious technical challenges, I did wonder how a site like this would handle some of the delicate issues around uploading pictures and/or 3D models of a woman's figure. How would this work? Would women be standing in their underwear in front of a webcam and rotating the body as prompted? Even if the collected data is a single texture or point cloud, it still could be considered a little...invasive? Not to mention the privacy issues.
Then use that information to construct a 3d model and apply clothing textures accurately to them
What would be really cool though, is an inflatable mannequin which could be inflated to near perfectly match that point cloud, which would then be clothed in the garment you want and you would receive a pic of your figure in that actual garment.
Logistics though? How do you (cheaply) mass produce mannequins with entirely different dimensions? 3D printer?
Also how would the clothing be fitted to the mannequin? Robots? Do we have a robot that can dress people yet?
That is awesome!!!
This is probably something Facebook or Google should do. They already have the users, if they had something functional like this they could get a significant cut of the global retail industry overnight.
Once you had that working, you could go about providing the same model for homes, interior and exterior. And buy Fedex.
Women already have a rough idea of how different sizes from different brands will fit them. Once they find another woman with a range of things that would also likely fit them well, they can have a reasonable assurance that their other clothes will be similarly sized and cut.
Once they find a few women of similar size, shape and taste, they can draw preferentially from that pool of users.
And how exactly do you propose the company profiting off their own buttons? (Items in a database they can create at will)
But the article was poorly written, so I am not sure if I was missing something.
Cheers to fewer underutilized dresses and belt sanders ending up in landfills.
Doesn't it bother any of you that she abandoned her entire Australian userbase?
'She says the site was so beloved in Australia that women still email her asking when it will be back up, “I’ve proven the model in Oz."' If it was proven in Australia, why shut it down? Did she need to shut it down for investors to take a look? Is there some YC criteria that wont let you have a successful product already?
Even with the variation, the bulk of them appear to be $40 or more, which seems like a lot to me. If one is shopping used, pretty dresses can be had for under $10 at a thrift store. It makes no sense to pay so much for a dress I can't even try on first!
I remember another site based around trading books that used a virtual currency, and a lot of the early adopters got free credits to help seed the system, while the late adopters could only get credits by sending books. It created an imbalance from the beginning.
Another issue is dealing with hoarders and retailers, which can damage the button system by putting too much virtual currency into the hands of only a few.
'need' is an interesting word to use in this context.
it's certainly a market need and it has real use; Humans make snap decisions on appearance.
If you're making a decision on someone over the course of fortnights on their clothing decisions, that is pretty much not a 'snap decision' by definition.
The issue I have is with the particular phrase - the linked article is talking about 'how women buy clothes', not 'how fashionistas buy clothes', and most women do not need (or even want) to buy new clothes every fortnight, let alone this idea of wear-once-and-toss-away. It's just the next impossible ideal of the fashion world.
I don't really have a problem with the product, I'm all for swap centres, I just have a problem with the language of need.
Anyway, within the scope of moderation systems, the HN system is in dire need of overhaul - it's terrible, and often leaves people annoyed and puzzled. Within the scope of clothing, people do not often need new clothes every week. A very, very small segment of people might be actually negatively affected by not having new clothing every week.
But honestly, the subtext of my comment is that this product just feeds into the impossible ideals of the fashion world. Fixing the moderation system on HN can be done by one coder in an afternoon. Fixing the multiple complex issues with body images is not so trite, and while not catastrophic in the traditional sense, certainly does have global consequences.
I don't know if she would like this service or not (a lot of it would, like many web apps, come down to how easy it was for her to use) but if she did, man oh man, would you ever have a dedicated customer. And I bet that goes for thousands, if not millions, of other women. Great idea!!!
What it strikes me this business needs are small pre-paid boxes which will fit in US mailboxes or can be left for your mailman.
"People who said this item fit well, also said these items fit well..."
2) Look at this image to see the difference between different brands and their size 8 measurements.
http://www.nytimes.com/imagepages/2011/04/24/business/201104... via http://thesocietypages.org/graphicsociology/tag/womens-size-...
There's many variables to sizing. It wouldn't be impossible to create a database of measurements for every single piece of clothing.
And then there's the issue of shrinkage.
Those are irrelevant details. The basic form of input is "person A gives rating R to item B". This is precisely the formulation behind successful solutions to the netflix problem.
What I find a little disappointing is that the top 3 prizes for this Netflix-Prize-style competition total $7,000 in value. Is technical brainpower, even that of students, really worth that little? Maybe startups should band together to hold their own optimization contests if the 'market rates' are so low.
People saying that size numbers vary greatly are missing the point. If I say a particular piece fits me well (regardless of its listed size) and then a bunch of other people say it fits them well (regardless of its listed size), the site can then aggregate the other pieces of clothing that those people said fit them well to show me clothes that potentially fit me well (regardless of their listed sizes).
I would just call it data analysis, not crowd-sourcing
EDIT: On an unrealted note, the name "Forever 21" is incredibly sad.
Not sure this will work.
Obviously your experience is different. Happy to agree to disagree about the outcome, but lets not pretend like either of us have a definitive knowledge of all women.
However, it could be totally anonymized, with measurements being used only to filter what is displayed. Think of it like a dating site, where whether someone smokes, or is a vegetarian, or whatever can be used to gauge compatibility (such as it is). Nobody knows my search criteria on a dating site.
These dresses are equivalent to "found money."
Woman has a closet full of dresses. They're old, uninteresting, unwearable, etc. She posts them online, earns "buttons," and buys dresses that look like they might work. Some do, some don't. Net result: more dresses to wear.
The "buttons" part is even more clever, in that they're building a real ecosystem that doesn't center on cash. Also, I can see this working incredibly well for pregnant women; they only need their maternity clothes for 5-6 months or so.
Sizes in women's clothing have no objective standards at all (I do so envy men's pants, which come in 34x30 rather than Size 8 Slim Petite), but Kendall Farr opens her well-known book by suggesting that women break out the tape measure anyway. Measure yourself, and then measure anything you're thinking about trying on -- it saves time.
The idea that stands out most to me, is this is the sort of idea and company, few male hackers could make work.
Basically, my killer feature was piggy backing on Facebook. Women love to upload photos, the photos are usually of them looking good in their dresses, the social network is already there to provide viral growth. So basically all a user would have to do is select from photos they've already uploaded, and describe the dress. Then their friends could see who has what dress available and see what it looked like actually being worn on a dress model they personally know.
My model really only worked with immediate friends, and was based on borrowing. Monetization I imagined would come from advertising shoes, accessories, and other dresses etc. But who knows, maybe women are willing to buy dresses from their friends?
I think it can still be done though, so the best of luck to Nikki and all the other sites that are trying to solve this problem. It is a big market and someone is bound to find the right solution that will work for the masses.
What? They're trying to create a large finite closet? The universe is in no danger? Curse you, TechCrunch! Fooled again!
The user experience for this could get bad fast when a new user signs on to trade their month-old H&M+<insert expensive designer> collaboration and finds nothing but a bunch last summer's Forever 21 shirts with cheeky slogans. That doesn't look so good. It's like going to the video rental store ten years ago and having them say "Whoops, we're all out of that new release, but we've got a few copies of Groundhog Day!"
http://www.reddit.com/r/malefashionadvice/comments/gyg6z/i_g...
If I find a pair of pants or a shirt that fits me well and I enjoy, I will keep it and wear it repeatedly until it is too worn or stretched out.
In a lot of ways, that's the opposite problem form what this site is trying to solve.
I would like a suit service just like this.
http://en.wikipedia.org/wiki/Greshams_law#Application
I.e. in this case the danger would be that users have an incentive to glut the system with low quality items (defects, uncomfortable etc.).
There's always stuff you can do like implementing user reputation systems, etc.
You'd basically pay an up front cost, say $XXX and then you would be able to order a week's worth of clothes. Then you'd mail the clothes back for laundering before being sent out again.
It is really a half-baked idea and requires a sizable amount of logistics, etc. But it was an interesting concept to toss around.
There's also a huge boon to be found in having branches in both hemispheres. Instead of spending December paying the land tax on a warehouse full of shorts, just rent them out in Australia. Then, come July, send out the shorts in the US and the sweaters down under.
I'm not sure about the environmental impact. On the one hand, you have quite a bit of carbon footprint from shipping the clothes all around. On the other hand, the company would have a strong economic incentive to maintain the clothes. I'm betting a lot of clothes that are currently thrown away could be saved with proper cleaning and sewing skills. I'm now imagining reading the company's blog having less posts on Redis integration and far more on their new way to remove grass stains.
Honestly, given the sizing issues, this probably makes more sense for men's clothes, but that avoids competition with 99dresses.
Dammit. I'm going to be thinking about this all night.
I like the idea of having a somewhat curated selection of some top brands like Banana Republic, J.Crew, Gap, etc. and being able to choose a certain number of items to take out at one time. I'd pay upwards of $200/mo if it meant I had an endless rotation of new clothes to try on without having to keep. And if you like the garment? Feel free to keep it and pay retail.
The collaborative consumption model is disrupting industry after industry, and it's only a matter of time before someone gets it right in fashion. I would honestly consider starting this myself, but I just left a startup and joined the Goog less than a month ago--need some stability for awhile.
How about using the library system, having a certain amount of deposit on the account, if the user lost the clothes then the presumed value of the clothes will be deducted from her account?
Let's say every user of the system has karma points. Karma points cannot be bought with real money. They are gained mostly putting unused dresses in the system, but also reddit-style, via comments, reviews, helping people and generally being a good citizen of the community.
1) Let every user list the dresses they would like in order of preference.
2) Suppose that every dress has a list of favourite users, based on karma.
3) Assign a dress via Gale-Shapley algorithm[1] or a variation, generally solving the stable marriage problem.
The fundamental idea of reselling used clothing is really cool and needs an overhaul from the usual Goodwill and other thrift stores, but 99dresses seems targeted at a specific audience that quite literally only wears something a couple times before tiring of it. I'm not sure how compatible that is with constantly changing fashions and questionable clothing quality from some of these clothing stores... I have a lot of old non-designer dresses I could be selling on this site, but after a couple times they've all got small problems (I don't want to sell something I mended!) or are out of fashion by a year or two.
There must be a scare for potential users in not receiving real money for sales, but something which can collapse or fluctuate in value at any time (also, it is potentially almost valueless if for instance you move somewhere the website doesnt operate).
Maybe not all people are as skeptical as me, but I'd imagine that this model will prevent users from 'leaping in' by selling large amounts of clothes, (and thus ending up with a huge store of 'buttons' that they may or may not ever find a use for).
I am very curious as to what the founders see as the upsides to this model which overcome said downsides.
http://www.startupsmart.com.au/cash-flow/99dresses-shuts-dow...
... can both possibly refer to the same company. At the very least, the prior shut-down of 99dresses has been given a quite a thick layer of spin.
It also reminds me that I've been meaning to clean out my closet, maybe I'll try putting whatever the thrift store won't take up here.
It's still a good business though, because an online store can have a bigger selection and serve people in areas without exchange stores.
This market is probably all about getting the critical mass of users so you have supply.
Many sites do not implement view all or infinite scroll and, empirically, still manage to make quite a lot of money. One prominent example is Amazon.
More broadly, I would sincerely love it if HN did not try to play "Let's one-up the founders" on day 1 for every startup, with special attention paid to YC startups. All startups will have great big honking problems on day one. (They're more of the flavor "Nobody knows we exist" than "Our 100 products are paginated rather than on a single page.") The ones that succeed in solving those problems will, in the process, revisit almost every stupidly inconsequential implementation detail, in the same sort of painful, considered depth that all of us who have shipped software before know will eventually happen.
Wouldn't we be a happier, more productive community if the tone was less "I have found your flaw!" and more "That's a good start. You might consider adding X, Y, and Z to the roadmap. Those have previously had outcomes like X1, Y1, and Z1 when tried in circumstances X2, Y2, and Z2. X and Y can probably wait, but Z should be a fairly high priority because $EXPLANATION_OF_HOW_Z_PREVENTS_YOU_FROM_DYING."
More importantly, I think you are wrong to think "Nobody knows we exist" and getting the details right are somehow separate. I built a business by a thousand small fixes. I worked on getting the SEO, copy writing, UX, ads, etc. better every day. Then one day the conversion rate was high enough that buying ads became nicely profitable and word of mouth started to take effect. Details matter.
If a fashion site has an UX issue that makes it not fun to use, it is a real problem. There is no next button at the bottom of the list of items. It feels like the site is broken, but this such a trivial problem to fix.
[0]: http://en.wikipedia.org/wiki/List_of_fashion_designers#Unite...
That said starting with women and dresses and later expanding to guys, say for dress suits/etc.. wouldn't be a bad idea.
all kidding aside, I bet women are going to love this.