Online Fashion: 9 Innovations for Online Sizing

Lately, I’ve been trying to problem-solve  how to get accurate fittings online for a product like custom mens’ shirts. Here are 9 possible innovations in determining sizing when shopping for clothes online and the advice I received from a handful of clever friends.

1. Custom-made balloons in the shape of fitting manequins.
Customers place them inside an existing garment until they represent the fit of their body shape. Customers then take a pressure reading and/or read from pre-printed measurements printed on the balloon.

2. Rubber shirt. Similar to the balloon concept.
But using a standardised tube of rubber that the customer stretches over their torso. The customer could measure the tension in the rubber using a set of spring scales. The advantage of these first two is that they would be very cheap to produce and post to a potential customer.

3. Piggy-back off the big guys.
Ask customers to fit themselves at a trusted, worldwide chain like Uniqlo and provide their ‘Uniqlo size’ with additional instructions — i.e. +2 inches of lengh on sleeves, +1 cm in neck, -1.5 cm across the shoulders.

4. Create a design database.
Create a database of common shirting brands and their sizing/shapes to allow users to specify their best fit based on their favourite existing shirt and the option to provide additional instructions.

5. Use local expertise.
Pay local tailors in each city a small fee to measure the customer and forward sizing figures. Could also give the customer gift card in the value of a standard tailor’s fee as part payment on their second shirt order.

6. A digital xerox.
Customer sends their best fitting shirt to one of a few regional hubs where their existing garment is digitally scanned and returned to them within a couple of days.

7. Downloadable measurement device.
Printable, custom measuring tape that could be cut out of an A4 sheet of paper, pre-marked with standard sizing. Customer uses the tape to measure themselves and selects the most appropriate size, as well as providing additional instructions.

8. 3D modelling at home.
3D modelling using a standard webcam, which is the scaled to lifesize (1:1). Something like this.

9. Copy cat.
Send your best fittng garment to Shanghai as a fitting guide and have it returned with your custom shirt, designed online.

Earlier this week, I sent these ideas to 10 of my cleverest (all more so than me) and closest friends and requested their thoughts on the matter. The focus wasn’t on what is possible. Rather what might be possible.

The most popular ideas were #4, #6 and #8. Mike Knapp combined an idea that we have discussed before—making a dummy shirt out of inexpensive materials—with the a the genius suggestion to embed tape measures in key areas of the dummy shirt (i.e. neck, cuffs, shoulders and back pleats) to allow customers to accurately adjust the sizing to their body before returning it for production. Sophie Hirst reminded me that despite many of these ideas being possible and even leading to a better fit, many of them require a greater investment in time and effort than the customer’s perceived value of the shirt (both emotionally and physically). James Avias made made the point that mastering the technology associated with #8 would transform the simple online store for custom mens’ shirts into the a tech start-up.

The major design issues I foresee are:

  • Time and cost of databasing a critical mass of mens shirting designs and converting them to the sample sizes held by a single manufacturer (#4).
  • Even if it is possible to simply scan the key panels/sillouette of a particular shirt, the shirt would still need to be sent to China (where I am living) until there was enough sales volume to determine where regional hubs should be established (#6).
  • Untested technology, which would require a lot of development—programming is not my area—and capital investment. I plan to test the capabilities of existing software further though and contact the respective developers for more information (#8).

Do you have ideas to add or opinions on those above? Please let me know your thoughts in the comments.

  • Patrick Searle


    I meant to send this to you by email but given it’s now online, here are my comments online 😉

    In short tho my criteria as to what entails a ‘good solution’ is the following:
    a) Be super easy for the potential customer / have no cost to the potential customer
    b) Be easy for you to run / be relatively low cost to run
    c) Be an asset that you can use for marketing (i.e. SOP’s Shoe Designer both creates WOM and also allows users to share their designs via Facebook/Twitter)

    So here goes:

    a) Fail – Would need to send ballons to customer = time = decreased conversion, too complicated for customer = hassle = decreased conversion
    b) Fail – You would have to manufacture every balloon but never receive them back
    c) Ok, but not perfect as has little online spreadability

    a) Fail – Same as solution one
    b) Fail – Same as solution one
    c) Same as solution one

    a) Fail – As customers would have to find a Uniqlo (or equivalent shirt), find a Uniqlo shop (and hence prob buy Uniqlo shirt), still not have this shirt ‘tailored’
    b) Win – No cost to you
    c) Fail – V.boring solution to this problem

    a) Win – This is cooler and actually the concept of a ‘brand size database’ I think makes more business sense than the custom shirt idea. i.e. Essentially you are trying to solve the sizing problem that comes with e-commerce sales…..Let’s talk more about this 😉
    b) OK – For this idea will be costly to continue to update this database
    c) Win – I think this brand size database will generate lots of buzz, is highly marketable see a)

    a) Fail – Too much hassle, too much risk customer would buy tailored shirt from this tailor
    b) OK – Little cost to you
    c) Fail – Would generate no online or offline buzz

    a) Fail – Customer should be able to do this themselves in their own Zerox, customers afraid off losing said piece of clothing
    b) Fail – Have to pay to/from postage costs, have to buy many Zerox machines in multiple countries, have to employ staff in multiple countries
    c) Fail – No online or offline buzz can be generated from this

    a) OK – Not perfect as I think customers will still struggle to accurately measure themselves, therefore give up = less conversions
    b) Win – Almost no cost to you
    c) Fail – No online or offline buzz can be generated from this

    a) Win – V.Easy to do, no cost to the client.
    b) Win – Tho initial software setup costs following on from this no cost/time needed on your part
    c) Win – People will talk about this/share their designs with people online

    a) Fail – Too much hassle on the part of the customer
    b) Fail – Too much cost in terms of postage/human resources on your size
    c) Fail – Not buzz worthy

    So to sum up I think solution 8 is the best. Though actually I think option 4 has the best future business viability….

  • Andy

    Thanks for the detailed analysis Pat. I really like your criteria. But I disagree with some of the fails. I’ll give an example.

    1) Balloons can be made for next to nothing. Granted (as Sophie mentioned), people may not go to the trouble of messing around with a balloon. But it’s definitely a cost effective solution. For marketing, I think it has potential as each balloon that is sent out could essentially be a piece of marketing collateral and you’ve instantly got rich data on that customer before they’ve even bought anything (i.e. address).

    It’s still not a WIN idea—most of them aren’t in their current form—but as I talk about in this post (, a slight tweak on one of the above ideas might take it from totally unbuzzworthy and inefficient to a workable solution.

    Let’s chat more offline about #4.