Monday, May 11, 2009

The Shoe Sizing Conundrum

Warning: A long post. If shoe sizing issues don't interest you don't read on.  :)

To custom make shoes, we need our customers to provide us accurate foot measurements. Telling us you're a size 8 doesn't work because different brands size shoes slightly differently. We also can't do what Zappos and other online shoe stores do - they suggest you order a couple of sizes and just ship back the ones that don't fit. We can't put our custom made shoes back on the warehouse shelf like a standard shoe retailer can.

To get sizing right, the suppliers we are speaking to require 3 measurements:

1. The length of the foot (at the longest point)

2. The width of the foot (at the widest point)

3. The circumference of the foot (around the widest point in 2 above)

In physical stores that custom make shoes this is generally done by having the customer stand on a sheet of paper, the store assistant traces around the foot then measures the length and width using a ruler. The assistant then uses a measuring tape to measure the circumference of the foot. Even when this is done, 5%-10% of shoes need adjustments made to get them right.

The problem with having people do this same process at home is that there is so much room for error, and 6mm in length is a full shoe size! Just using a small circumference pen versus a large circumference pen, let alone angling the pen in or out against the foot, could make the difference a whole shoe size which would equal a customer return.

Here's what we've come up with as options to measure shoe sizes:

1. Customers do as described above and we create a good YouTube video to explain how to measure your foot accurately. If you don't have a measuring tape we'd have one you could print then cut out.

Pros - fairly simple.

Cons - prone to error. An assistant in a shoe store does this a lot so they know all the tricks to getting it right. We can't teach people at home to do it this well in a short video.

2. We do a slight variation on 1 above. Instead of drawing around your foot then measuring the length and width, you put your foot on a piece of paper, in a corner, on a hard floor and you mark the longest and widest points of your foot then measure them.
Pros - simple and reasonably accurate.

Cons - we're not sure of the accuracy of this compared to 3, 4 and 5 below.

3. We create a printable PDF document with measurements on it. You put it in a corner on a hard floor, then place your foot in the corner and measure the length and width of the foot on the document using the lines marked on it. The PDF document would have a measuring tape to cut out to measure the circumference of your foot. We'd explain how to do all this on a YouTube video.
Pros - quite accurate from our testing.

Cons - You'd have to:

1. Print two sheets of paper and tape them together in the exact right place (20% of women have feet too big for a standard page);

2. Ensure your printer hasn't scaled down the image which would change all the measurements;

3. Accurately fold along the edges of the paper (different printers will start printing at different points on the page);

4. Print and cut out a tape measure;

5. Do the measuring.

That's a lot of slow, potentially frustrating steps.

4. When someone signs up on the site, we post them a document as in 3 above.
Pros - accurate and simple.

Cons - posting the letter would take a couple of days. We'd lose customers while they wait for the sizing document to arrive.

5. We create a database of brands and shoe sizes. You tell us you're a size 9 in Jimmy Choo's. We match that in our database and can make you an equivalent size. (Thanks to Isis for this suggestion)
Pros - simple for the customer.

Cons - 

1. Difficult to build this database;

2. For some brands you might be a 9 in one style (Eg. a platform) and a 9.5 in another style (Eg. a flat).

Sizing is tough. I think on our next trip to Hong Kong we're going to have to nut this out with the suppliers and test, test and re-test the different options.

Does anyone have any other thoughts on what we could do?
(Photo Credit)


  1. I think #5 is really the best you could do if you could get a good DB together. Even if you offer people some different ways to find their size you could have an option such as: "Go to a store and try X, Y or Z shoe and tell us what size you were". If you're looking for return users then the hassle of testing once / getting good sizes lots is worth it I think.

    The hard thing even once you have a good foot size is to measure the shoes that you sell and make. As well as having size and comparative size (compared to your "default shoe) you'd need some good fuzzy information like "a tight fit size 10 shoe, hard around the front with soft cup at the back". Kind of like a wine explanation. Even if you don't know exactly what it means hopefully it will give you a "feeling".

  2. These all seem like pretty good ideas. One thing that would be amazing (and new) would be to actually have a foot mold of all your customers, to make sure shoes fit before you send them. If you could figure out an easy way for customers to take/send their foot mold to you, it would be so, so, so cool. I'm imagining that you send them a shoe-box sized mold, they heat it up, and press their foot into it, and then it hardens as it cools. (Like an easier version of: That technology alone could definitely encourage me to buy at least one pair of women's shoes from you (note: I am not a woman) just to take/have a mold of my foot. In the future, you could actually show these people what shoes on _their_ feet might look like. Sah-weeeet!

    The other thing you could do would be to have some "tester" sizes for people that aren't entirely sure of their foot measurements, or confident in their measuring skills. They'd give you their size, and you'd ship an ugly tester to them to see how it fit, which I assume would be cheaper and could be reused many times. This would require more patience on the customer's part, but maybe you reward them with a discount or other incentive to get the sizing just right.

    After all, you only have to solve this problem once per customer, right?

  3. How about having them stand in the corner as in #3 but instead of doing any measurement themselves, they just put a standard-sized object (e.g. a dollar bill) next to the foot, take a digital picture, and send it to you. You could then do a little image analysis to determine the size of the foot.

  4. I think the beauty of the internet is that it's highly dynamic. You've also got the ability to have a closed beta to test some of these options.

    Why not offer several of the above options and see what the uptake is?
    Create video tutorials for each method
    Create a printable measuring tape (so long as print scaling isn't going to effect this)
    Experiment with shoe molds that are posted out.
    Develop a brand comparison chart (though I would be surprised if some version of this doesn't already exist).

    I think you're on the right track when you say you're going to speak to your suppliers - they've no doubt been going through this before and have some of their own ideas.


  5. Tarwin - good thoughts. Cheers! The database idea I think has a lot of potential. The difficulty will be building it up - particularly if we want to service all countries. That's a lot of brands we'd have to get data for. It's going to be a challenge, but something that has a lot of potential.

    Cam - love the mold idea. Fox and I have been talking more about this today. Could be a real winner if we can find a good mold that was easy for the customer. (And I have no doubt you'll buy at least one pair from us! hehe :)

    Ben - excellent idea! Someone mentioned this to us as well today. We were thinking maybe it could be turned into an iPhone app. You place a ruler on the ground, and then move the phone around your foot. Quite tricky to do, but it would be golden if you could pull it off.

    Ned - you've hit the nail on the head there. I think we should definitely offer a few options to begin with, and then see what works. Hopefully though it won't prove too costly for us while we're getting started. Too many returns could bury us.

  6. Rulers can be tricky since you have to discern the marks on them unless they're precisely a known length (also they're not exactly ubiquitous). Detecting the outside dimensions of a known object is easier. There's an iphone app called RulerPhone that measures things using a credit card for scale (but I'd recommend at least having the option to upload a photo instead of an iphone app for widest availability).

  7. Hmmmm. Interesting! I hadn't heard of that app.

    After reading some more about it though, I don't know if it will work.

    According to the download instruction for RulerPhone, it's accurate to within half an inch (12.7mm) at less than 3ft away. That's two shoe sizes of potential error. :(

    I guess with higher resolution photography, maybe we could improve that. So maybe an iPhone app isn't the way to go - but the customers take a photo with their digital cameras, and we could then semi-manually perform of at least confirm the calculations...

    Something to mull over. Thanks Ben! :)

  8. Hey guys,

    There is no surprise that this topic directly follows the returns chat, as they two are dynamically related. The more you invest in getting the right shoes (style, fit, colour, look etc) to the right person (reliable delivery), the lower your returns. I suppose there is a formula somewhere off in the distance where you can work out your volatility to returns versus cost of getting it right first time.

    One thing that could work in addition to your measuring methods, are a few tutorials. For simiplicity sake, lets say you've got 3 diff suppliers. Well you get Mary to model shoes from supplier one and explain that she is normally a 10 as she has wide feet but these are...etc. I reckon as a girl (and I'm not) that I'd know I had wide feet and preferred certain fits, so another lass telling me in a youtube testimonial that she has the same issue and therefore went a size up, would be great!

  9. oh, and another thought from the science (vastly underused!) part of my brain. What about a control sample? Is there a shoe shop that exists in the markets you want to sell, where you can say "if you are a size 10 in this shoe, then you'll need an 9 in Shoe A but a 10 in Shoe B?

    Reduce the variability and manage the risk.

  10. Hey Dom, good thoughts there. We could do a simpler version of the shoe size database and just do it with a few major brands.


  12. Dom, those links are brilliant, thanks! Some very detailed descriptions of how to measure feet, I like it!

  13. i think these have all been thought of before but anyway..

    1: mouldable (moldable?) or adjustable slippers. these would measure length and height of both feet. unfortunately they may not have been invented yet! ;(
    1a: maybe a cotton footbed they can trace around with a marker on the bottom, with velcro straps for the top.

    2: I like the impression foam thing mentioned in Cam's link (i assume that people who cant measure their feet properly cant make a plaster cast either). one supplier from google:,2246.html
    you could add a strap/s to the top they wrap over their foot.

    I reckon if you could send these out for new customers (cos its repeat business you are after right?) and if they only cost ~$5 with return postage...

    3: get them to send in an old pair of similar style that fit, with postage cost receipt. if they are anything like Lisa, they will have about 200 old pairs of shoes they could part with. you discount on first order the postage (and send the shoes back if they want).

    The photo idea is brilliant too, I think. put their foot next to a ruler. a photo for length/width and another for height. so simple, so easy, no stuffing around with postage.

    have you got any orthopedic doctor/supplier friends?

    best of luck mate!

  14. Thanks Aaron, good thoughts on adding a strap to the foot impression foam, that could help. We spoke to our supplier about the foot impression foam though and they didn't like the idea. I'm on my way to China now for a couple of weeks and I'll be speaking to them in a lot more detail about ways to solve sizing. For the moment though, they like the idea of posting an old shoe and that's a good thought to have the customer include the postage receipt and we take that off the price. Cheers. :D