I dropped over 40 grand when I could have spent $100

“You dropped a 150 grand on a fucking education that you could’ve got a $1.50 in late charges at the public library” – Good Will Hunting (1997)

I love that quote. It not only speaks to the accessibility of knowledge, it also speaks to the importance of being efficient with your learnings.

I just recently shut down Rewire Attire, the exclusive high end apparel marketplace. We hand picked and recruited high end fashion designers to list their designs for sale. We marketed and sold those designs. We arranged shipment and accepted payment. Designers packaged their orders and dropped them off for shipment.

One of our biggest selling features for shoppers was our “perfect fit” technology. The pitch was to get away from inaccurate generic sizing terms like “small”, “medium” and “large” and move to a system based on body measurements. While it’s impractical to use body measurements in a traditional bricks and mortar retail environment, the internet is well suited to log a shoppers body measurements and compare those measurements to a catalogue of fashion designs – “have designs try you on instead”.

Even after shutting down the business, writing about it still gets me excited. I still feel the future of fashion retail (or clothes shopping) involves an improvement on sizing with the help of the internet (and data).

Here’s the problem. I’m not the target audience. The target audience for Rewire Attire, the high end fashion marketplace, was women with enough money to afford high end fashion and that are young enough to be comfortable shopping for clothes on the internet. It turns out that these women are not comfortable sharing their body measurements, overwhelmingly so, despite the benefits.

I found this out by spending $100 on a Google consumer survey:


…after already investing over a year of my time and over $40,000 developing the business.

Even though I’d listened Eric Reis speak at Stanford (via podcast…thanks Stanford for sharing). And even though I read Steve Blank‘s The Four Steps to the Epiphany. I was still inefficient with my learnings.

Needless to say, I’m now an even bigger proponent of The Lean Startup.



61 thoughts on “I dropped over 40 grand when I could have spent $100

  1. Would you consider reconfiguring Rewire Attire for the big and/or tall man? As a man of larger carriage (fat), finding well-made, stylish clothes of natural fibers (read: not polyester/rayon) that fit well is a challenge. I would have no problem sharing my measurements, and, based on the success of made-to-measure sites like Indochino, other men are willing to share, too.

    • Hey Jeff, thanks for the comment. It’s an interesting idea. The other piece of RA’s business model, which I didn’t highlight in the blog post was that it was an exchange marketplace. We didn’t design or manufacture the items we sold. We recruited designers that did. We had more success finding women’s high end fashion designers, which probably speaks to the problem that both you and I have (I’m a tall man).

    • I agree… I am tall and fairly lean. Almost nothing fits well. When I look at amazon and filter to my size, it often comes down to a selection of fewer than 3 items.

    • 6’5″, 265lbs. I have a hell of a time finding anything that fits that looks decent. Big and tall departments for some reason assume everyone big and tall is old and blind. I looked into https://www.trunkclub.com – love the idea but they don’t carry anything in my size. I’ve given up on online shopping. I would LOVE to be able to upload my measurements and be offered stylish clothing that actually fit.

    • Likewise. I’m not that big any more, but I’m a slightly odd shape (slightly shorter arms, chubby thighs) which means I have to try cloths on, buying online only works for t-shirts, and even then it’s a bit of a lottery.

      Being able to upload detailed measurements would be a no-brainer for me if it meant I could actually buy stuff online that would fit.

    • There was one thing I neglected to mention in the blog post, that I should mention now. We ran a contest (prior to the survey) that required contestants to submit their body measurements. We wanted shoppers to see the power of submitting their measurements. We threw quite a bit of money and traffic and the contest and we saw almost no engagement. My thinking was, if we can’t get shoppers to submit their measurements for a chance to win a free high end fashion design, we might have a problem (let’s do a survey!)

      • How do you know it just wasn’t a poorly communicated email campaign? The idea sounds great to me. I believe you had a trust issue not an idea issue. I could image my wife thinking “Now why the hell am I giving these guys my measurements?” One email just isn’t going to do it; she’d need to be educated, wooed, then convinced. It’d be a process.

      • You’re likely right, but even then, it’s an expensive process. There was a lot about RA, outside of sizing/measurements that required educating, wooing and convincing. We worked with emerging designers (read unknown designers) because they were most open to the exchange marketplace model. RA was itself an unknown brand. Without some outside investment, we just didn’t have the juice to muscle through that challenge.

  2. Hey Tim,

    I don’t know if you’ve hit a roadblock. Innovation is all about improving existing offerings but also ‘plowing new snow’. You’ve stumbled across a potentially new market and I think your next step is to figure out what it takes to make your ideal customer aware of their problem and what offering could overcome the inertia.

    Just a couple of ideas, what if you used social proof to sell these ladies on how great the clothes can fit.

    Or what if you offered a concierge service that would help them get fitted? Have you ever walked into a Brooks Brothers and had them take your measurements? Boy does that feel good 🙂

    Just a few thoughts.


  3. Way too little data to make any decisions. You’d need hundreds more responses to make any confident decisions based on this data. In fact Google is even telling you as much by giving you confidence intervals that are very wide.

  4. Ah heck, I thought this was going to be a blog-post on how somebody spent $40,000.00 on a standard Linux supercomputer when they could have spent $100 on the new Parallela supercomputer that was crowd-funded on Kickstarter. But it was still an interesting read. Thanx.

  5. I had an idea the other day for a cross between groupon and kickstarter for new designers. you highlight the designers stuff in a email if it reaches a certain threshold of people that want it, you make it and send it out. Don’t have any desire to work in fashion so, I would never do it.

  6. I’m not a statistician, but with only 8 or 9 responses to each question, I’m not sure that you can make an argument that the data is statistically significant. I like the general idea, but you probably need to increase your sample size significantly in order to draw any conclusions.

    You probably also want to take the respondent’s location into account, as someone earning $23k/year in New York or LA has very different buying habits as someone earning that same amount in other locations.

    The danger is (obviously) that using poor-quality data can lead to you incorrect insights and down the wrong road.

  7. I am tired of hearing this story. People need to stop wasting time and money on building niche products for audiences that they ultimately know nothing about. I cant tell you how many shopping sites I visit that are so obviously designed and created by people (mostly men for women) who know nothing about selling to their target demographic. I am glad you have reevaluated at this point and not further down the road as many others have.

  8. Is there a way you could figure out their rough measurements anonymously? What I’m thinking is maybe having them tell you about a brand of clothing that currently fits them. Like if they say, “I’ve got a size 4 Vera Wang gown that fits perfectly!” then would you have some idea of how that’s sized? Could you then make a suggestion for buying something new? And once they’ve bought a couple pieces new and rated them as fitting properly, could you then know approximately what fits them and offer suggestions based on that?

  9. Given that I have a gigantic head….. glasses, sunglasses, hats, du-rags or whatever other accessory meant to be worn on the head / face never fit.

    There’s your marketplace and your pivot. I’ll be your #1 customer.

    • I think it’s fascinating that we’re seeing a common theme emerge here: If FIT is a problem for you, if you’re a “special size” in any way, then you would absolutely not hesitate to submit measurements.

      But if you’re “regular” (whatever that means) then it’s a pain in the ass and you won’t do it.

      I’m actually thrilled to see this trend emerging and the refrain being picked up by so many, because it was my own hypothesis back in August when I started my business.

      It’s now being somewhat validated for free thanks to this post. #magic

  10. I think the issue with women are numbers. Women are terrified to measure them selves and be disappointed, or have other people find out that they are fat. Also, women are their own worst critic. Before you give up on this idea, try this out, reword your pitch.

    1. Never talk about measurements, talk about sizes.
    2. Make custom tape measures with letters instead of numbers and your company logo on the side. Have the letters start out fairly close and then spase out as you getting closer to Z. Something like A-B–C—-D and so on. The idea is, a girl with a slim figure and her slightly chubby friend would only be a few letters appart. Sell this idea as the new high tech method of buying clothes, which can only work with the special measurements.

    Don’t think this will work? Try subbmitting another google servaym but word the qustions like this:

    1. Do you find it dificult to buy clothes online that fits you perfectly due to the confusing or inacurate measurements that most murchants provide?
    2. Would you consider using a new and unique sizing system that takes into account your bust, waste, hips, and hight measurements to find the perfect designer dress custom tailered for you?

    Spend another $100 before you throw away $40k, run the survay.

  11. Eric Ries and the Lean Startup methodology wouldn’t advise to do a survey to see you’re not choosing the right idea or market either. You have to test whether people would actually use or buy your product, not what they say they would do via survey. Many people might say they would pay for something in a survey when in reality they’d probably hit your landing page, see a sign up form, and then leave.

    • While I didn’t mention this in the blog post, it’s worth noting that we had a operational site and were not seeing the engagement we had hoped. We ran the survey to test/support what we were seeing in our logs.

  12. Tim, I think I have an interesting approach to accomplish what you’re looking to do (data-driven fit for clothing) without having women do something they’re uncomfortable with (“would you give out your body measurements? NO”).

    For everyone who says the dataset is too small to draw the conclusion that Tim drew: I’ll bet a Starbucks drink that if you grow the sample size 100x or even 10,000x, the effect will be linear and the results will be the same.

    Email me?

    • They could shop anonymously….but if they wanted to place an order or save their measurements for another time, we needed an email address.

  13. While I absolutely understand your point here–and wholeheartedly subscribe to lean methodologies!–I believe you’ve overlooked something of great importance:
    Your survey question was worded in a way that’s potentially off-putting to female consumers.
    Secondarily, I would like to hear more about HOW you were trying to create and offer that perfect fit.
    By asking busy, high-income women to take time to measure themselves (which is a wildly unplayable experience, just FYI) you’ve asked them to trade a very precious commodity: time. Plus add in discomfort, then a possible trust issue with the “security” of that data.

    Also: the richest women are over 35.

    The truth is, women who wear sizes 0-12 just don’t find fit to be an insurmountable obstacle. Meanwhile, Plus Size women (size 14+), who actually make up 67% of the US population and spend $18B a year on apparel, DO.
    So, at AbbeyPost we’re solving that problem. Partly by creating a “perfect fit” tool. The results are almost diametrically opposite yours.

    Just some food for thought on product-market fit.

    • Tim, I hope you’re starting to realize that the problem was not your failure to do a survey but rather your decision to jump into a high-cost market segment and offer an overserved demographic something they can’t imagine needing. I’m an ultra-busy high-income *fat* female just over 35. I can spend hours shopping and come home empty-handed, and my time is too scarce to bother trying very often. Like the tall and big-tall guys who have already commented, I would gladly hand over my measurements if a credible number of measurements were collected, confidentiality was ensured, and it could result in a dependable stream of great-fitting clothes.

      Your survey would have saved you tens of thousands of dollars. A little forethought and a slightly longer survey might have earned you much more than that.

      • Great comment, Karen. And please check out AbbeyPost.com (yes, it’s a shameless plug, but we’re actually working on a solution for the exact problem you describe. And in the meantime, we have great stuff for sale.).

        Hey Tim, I can’t help but notice that you have replied to many comments here but not any from women (even though we’ve been highly specific in our feedback). Um. We’re supposedly your target market. What’s going on?

  14. You’re making a common mistake that many with the power of running a survey make: You’re coming to a conclusion based on data that’s not at all statistically significant. Your response pool is so tiny, it’s not indicative of much of anything, so you can consider the results invalid.

    Furthermore, you spent $40k building a technology and business. Did it work out? No. But I bet you gained valuable experience, made new friends/contacts, and learned something about yourself. Is that worth $40k? Almost certainly.

  15. You know how there are all these little apps to ‘build a website’ but the outcome is nowhere close to someone who actually knows what they’re doing? So it is with self-serve research.

    When dealing with survey data you must consider not only the overall sample size, but the sample size within each ‘cell’ (in this case, age * income).

    Your population of interest was women aged 25-34 in the higher income bands, of which there were a total of 2 – TWO – respondents. It’s beyond ‘small’ – it’s almost non-existent. Furthermore even amongst your female 25-34 sample of 45, application of weighting is not advisable because the portion of your sample that is being weighted up might be too small in itself – and weighting reduces your effective sample size even further.

    At the end of the day research is like anything else – if it’s worth doing, it’s worth doing well.

    Thanks for sharing, though; designing products for a market that *is not you* is an tricky problem and there is a lot to think about. All the best for your next endeavour.

      • Cynthia, no, I don’t work with Upcload. I have a degree in apparel design and am interesting in sizing and fit issues. Upcload has an interesting model for collecting measurements.

  16. I wouldn’t give you my $100 to invest. Much less $40k. You’ve demonstrated that you’re willing to spend money on ideas that you know nothing about. $40k down the drain on a business without knowing anything about your target market and then shutting down that entire investment based on a $100 survey that only had 17 responses. Statistics 101 isn’t even required to understand that you don’t have enough data on which to base a smart decision.

    That being said, I think there is something to be said for doing some relatively cheap market research using tools like Google Consumer Surveys before investing large amounts of money in a business idea.

  17. If you asked most people if they would be happy sharing their private photographs with strangers on the Internet, you’d have got an overwhelming no. Especially before 2004. Yet, Facebook found a sweet spot and exploited it.

    The question (you asked about giving up your size info) is bound to inform, but then you have to choose how you apply that knowledge to create an incentive for your target market to want you product. You’re innovating, so they cannot know the benefits unless there’s some social proof.

    If an influential person raved about your service, would that sway people to respond differently? I suspect so.

    My favourite example is this: what if I told you to roll up some leaves in paper, put that in your mouth and light fire to it? Sounds mad, but the social pull brings many people to do something many others find disgusting.

    Lean is good. But you have to keep agitating to get to the right ‘fit’. Not give up. The fail fast should be for a feature or assumption. Not for your whole vision.

  18. FYI – 5’11, 215lbs, 28inch chest, 34inch waist. impossible to find clothes / suits that fit. Always end up with jeans that are way too tight in the thighs or way too baggy. To get suits, usually end up getting pants that are 44 inch waist and having them taken in, if they even can.

  19. may be its not a site but an xbox360 game. you could use kinect to measure, data is local, never leaves the home xbox, and make true personalized shopping exp. and facebook shares of good fits, you know.

  20. oops. You did a survey instead of watching – Etsy.com did a billion dollars in sales of hand made clothing etc last year and every single one of the 28 million purchasers would have submitted their personal details. For the record if you asked ME, I would say “no, I won’t submit my measurements”. But I do if it’s worth it…
    I remember surveys around 1997 asking people if they would submit their credit card details online. The answer as always is “no.. until I find a good reason to… “. hello eBay. 🙂

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s