Imagine you are a pattern maker. You have a great design all worked out. But what sizes should you include in your pattern?
You might start by looking at what other pattern makers are doing. Then you might look at what standard size ranges are being used by ready-to-wear clothing manufacturers and large commercial pattern makers. But a small pattern maker can’t always afford to make every pattern in every size, so how do you choose?
A Look at the Numbers
The body weight of adult women follows what is called a standard distribution. If you charted out the number of people by their body weight, it would make a curve. There would be a lot of people in the middle of the curve (around the average weight), but not many people out at the ends of the curve (with very low or high weights).
According to SizeUSA, as of 2004, the average bust size in the US was 40″. That translates to roughly a size EUR 44, or US pattern size 18. That’s the centre of the curve.
Indie Pattern Makers
What many independent pattern companies choose to do, is offer their patterns in sizes 6 to 18 (or average). If you chart that out, you can examine what is called “the area under the curve” and use this to calculate the percentage of the population in that range.
For independent pattern companies that offer their patterns in sizes 6 to 18 (the green area under the curve), that area is about 43% of the total population. That means that if you make a pattern available in that size range, only 43% of the general population can use your patterns.
That’s not a huge number, so you might think about extending your range of sizes. But in what direction? If you look at the chart you can see that by extending the range down, say two sizes, you make the area under the curve just a little larger, but if you extend the range up by the same number of sizes, you make that area much larger. In other words, you can find more buyers by adding sizes nearest the centre of the curve (size 18), then at the ends.
Big 4 Pattern Makers
That’s what commercial pattern companies do for their regular pattern range. Often, they offer patterns in sizes 6 to 22. With that range, about 63% of the population can use their regular patterns. They’ve just increased their population of potential buyers by about half. But instead of 7 sizes their range includes 9 sizes.
An Ideal Limited Pattern Range
When you produce a pattern, there is an additional cost for producing and grading each size. Not every pattern maker can add extra sizes. So lets assume that as an independent pattern maker, you can only afford to draft seven sizes. Which should they be? In order to include the largest number of potential buyers, you should centre your seven sizes around the average size. If you include sizes 12 t0 24, 56% of the general population can use your patterns. In other words, by shifting the range of patterns you are providing up by a couple sizes, you can make more money at no extra cost.
Our average weight has been increasing over the past several decades. In the 1940s and 1950s, when standard clothing sizes were developed, the average bust measurement was 35″, about a size 12-14.
If you look at the Big4 pattern makers, this is the centre of their current range of sizes. Most likely, they originally organized their pattern size range to centre around the average size (in the 1940s and 1950s), in order to maximize profits. But since then, our bodies have grown out of that ideal range, and the Big 4 pattern companies have not adjusted.
If you look at the small range that many small independent pattern companies are using (6-18), they are also centring their size range around this outdated average, just like the Big4 pattern companies. But they don’t have the historical reason to do so.
A Sample Extended Pattern Range
And what would happen if you matched the size range on both sides of the average? What would happen if you offered sizes 6 to 18, but then also offered sizes 18 to 30? You’re patterns could be used by 86% of the population. You would double your population of potential buyers (and presumably your profits). Of course now you would also have to draft an extra 6 sizes. And of course, you can always extend that range even further.
And would that extra segment of the population buy your patterns? It’s hard to predict. But assuming the patterns were well drafted, and fashionable, you would be adding a population of potential buyers that are underserved by both large commercial and small independent pattern makers, but also by the ready-to-wear clothing industry.
Statistical Caveats
This model of pattern sizing is approximate and depends on a number of assumptions.
Human body weight follows an approximately normal distribution, so it’s logical to assume that clothing sizes do as well, but they might not. That data has been compiled by private companies, but it isn’t publicly available.
This model assumes that clothing sizes vary by equal amounts along the curve, that measurements like bust size and body weight are perfectly correlated, but that isn’t necessarily the case.
Nevertheless, it’s logical to assume that the model would be a good approximation of size distribution in the population and the calculations shown here would hold up.
So what sizes should you include in your pattern?
Based on time and budget constraints, determine how many sizes you can draft, test, and print. Then centre those sizes around the average, size 18 (EUR 44 or L with bust 40″), with an equal number of sizes above and below that point. (And yes, size large, is in fact average).
Of course, this takes into account only cold hard economics of choosing a size range. There are many reasons, besides profits, that would lead to different choices. In a perfect world, all patterns would come in all sizes – and they would all look amazing! (perfect world right?)
Your thoughts are welcome in the comments!
The Finished Garment 
You are awesome. Thanks for this. I needed to hear I was average today.
LikeLike
Glad to help!I sometimes like to be average too.
LikeLike
I love this! As an AP Statistics teacher and avid dressmaker, this post really spoke to me! I may even use it in a future lesson about Normal Distributions, if that’s ok with you?
LikeLike
Sure! I used to teach stats too. Just please include some way to give me credit for the work I did (like a link back).
LikeLike
I love sewing scientists! 😀
LikeLike
Me too.
LikeLike
Thank you for this! A very clear point. One thing that is missing: indie pattern makers sell internationally. I know for example that 40% of my readers is American. Are these numbers also valid for Europe, Australia, etc,?
LikeLike
It’s hard to get data for that. But SizeUK did a similar study in the UK in 2004 and found the average bust measurement there was 38 (one size smaller than in the US).
It’s hard to find measurement data, but if you compare body weight you can make some educated guesses.
The average body weight of American women, in 2010, was 166 pounds. In other countries: 162 Australia, 161 UK, 160 Canada, 158 Germany, 154 Spain, 147 France. In 1953, when the original “regular” size charts were developed and the average size was 12-14, the average weight of women in the US was 134 pounds. In Canada, it was 135 pounds.
LikeLike
Thanks for the additional data. To add, I actually use two blocks. One has a 34.5″ bust, the other one a 44″ bust. After doing research on how women increase in size proportionally I just couldn’t ignore that. I hope it will work out and be worth the extra work! My first pattern isn’t out yet so we’ll see 🙂
LikeLike
Shannon, you indeed make a very good case. Going to look into how much work/time/money it will take to increase my size range going forward….
LikeLike
Awesome.
LikeLike
Shannon, this is an great topic! I would like to say however that there is in fact added cost to developing larger sizes. In order to increase a size range properly beyond sizes 16 or 18, the pattern style needs to be redeveloped for a different body type. This is because as we gain weight we don’t gain it proportionately. In other words our body morphs and we change our shape. In addition, there are several shapes we take on as we gain weight. That is, some of us are apples or pears for instance. These differences in shape require different pattern shapes. Gerry Cooklin wrote a book precisely on this:http://www.amazon.com/Master-Patterns-Grading-Womens-Outsizes/dp/0632039159/ref=la_B001HD3SG2_1_5?s=books&ie=UTF8&qid=1403713639&sr=1-5
In any case, I think developing patterns properly for larger sizes would be fantastic. Every woman wants to feel beautiful in their clothes and proper fit is at the base of this in my opinion.
LikeLike
It’s true that grading takes skill – that’s why it’s a profession, and not a couple of buttons in Illustrator! But increasing the range is especially difficult, if the best matched pattern block is size 12-14, when the population centres around 18.
People are using an outdated, smaller-than average block and calling it “regular”, and then being forced to produce another larger block and calling it “plus-size” to include more people, when really, they should adjust the basic “regular” block to be centred around the true average, like it was when that “regular” block was first developed and truly captured the average human female form.
Maybe if we started at 18, pattern grading would be easier.
LikeLike
Grading is in actuality is the “easy” part. With computer systems now, it takes very little time and skill. The real skill is in developing a great fitting pattern that is based on the mid size of the entire size range, and you are absolutely correct in saying that adjusting the base size would solve many issues. I maintain that in order to reach 86% of the market would still require separate sets of patterns due to body shape differences throughout the size range.
LikeLike
Yes, you might need two blocks if you want to cover 86% of the market. I’d guess that the pattern makers that go to size 20 or 22 are just extending their existing block a bit? And at least some of the pattern makers that go out to that 86% range are using two blocks? That 86% is probably an ambitious range for many indie pattern makers. But, the payoff for shifting the curve, even by just a little, is still pretty substantial.
Thanks for bringing your expertise to the discussion!
LikeLike
In most cases the styles are just graded using one base size pattern. I believe this is why most garments don’t fit those that land in the upper or lower extremes of the size range. Consumers end up fiddling with the fit themselves or just making do with the result. This of course will happily work for some but not for all. If I just had a few extra hands I would definitely add this to my offering! It would definitely make for greater sales and happy customers. 🙂
Thanks for covering this topic and a really great discussion.
LikeLike
I have heard this argument before about differences in shape being the reason that pattern companies don’t grade up, but it just makes no sense to me. There are plenty of size 4 women who are apple shaped or pear shaped and they learn, just as a size 20 woman does, to make the adjustments necessary to fit their shape.
LikeLike
The reason companies don’t “just grade up” is because the further you grade from the base size, the more distorted the pattern becomes. (By base size I mean the middle size of the size range.) As Shannon mentioned before, changing the base size would make all the difference here. But this creates the need for different blocks for each size range again.
I agree that home sewers will learn the adjustments they need to make for their shape no matter what size they are but for the sake of creating a customer following I for one would like to provide a product where the pattern corrections are minimal to none for my customer. Perhaps this is a lofty, unachievable goal but I do keep this in mind as I develop the patterns.
LikeLike
I think the now defunct Stretch Pattern School website’s articles on sizing demographic & women’s sizing table are must reads.
Stuart’s observations are at least based on decades of experience in the garment industry & over 27,000 sets of measurement data he collected through his website. So I have more confidence in his observations than a lot our sewist speculations on how sizing should be done. (Though our needs & desire for better fitting patterns in our sizes are totally valid as well.)
From what I can gather, it’s not as simple as we’d like to imagine; and Alexandra is totally right that once you expand into the larger sizes you may need multiple blocks for fitted styles.
Personally, I’d rather indie pattern companies offer downloadables (hopefully cheaper for them to produce) targeting very specific markets (shapes as well as sizes) than trying to emulate the big pattern brands in catering for all & pleasing very few. Then you can trust that what you buy, if you pick the right brands, will be flattering without much alterations. (Hence my recent idea for a pattern brand matchmaking site. Though now I’m thinking it’s too much hard work for this sewist to take on }:-)
LikeLike
Hi Pia,
Stuarts collection of measurement data is quite amazing. However, the validity of the data is lost to a great extent when consumers measure themselves. There is no way of knowing if the measurements are accurate.
There are scientifically conducted measurement data available through ASTM: http://www.astm.org/BOOKSTORE/COMPS/APPAREL13.htm. But any measurement table/data does not adequately take into account “shape”. For instance you can have 3 people with the same bust measurement and also have three different body shapes that correspond to that measurement.
The dressform company Alvanon is doing body scans to collect the data for their business-http://www.alvanon.com/fit-movement/alvaprojects/-this to me would be the most reliable sizing information and it certainly costs to acquire it.
I love your idea for a brand matchmaking site. I think it would be an advantage to both independent patternmakers and the consumers they are looking to reach.
LikeLike
Part of the reason the older data was used for so long was that it was based on measurements taken by the government of the population. And there were two wars and two drafts, which meant that a huge number of people were measured all at once, but in the same consistent way. But yes, self-reported measurements are a poor substitute.
LikeLike
I wonder though for pattern brands whether they need to account for the fact that pattern consumers will be measuring themselves. We’re not going to have government or body scanning companies measuring us before selecting our pattern sizes. If the brand’s sizing is based on accurate standard measurements but your consumers are choosing based on their own inaccurate measurements then they may end up choosing the wrong size anyway.
Perhaps better instruction for measurements should be provided. For example, I never know whether I should have breathed in or out when I take measurements affected by breathing (chest, bust, waist). (I do like the fact that Stuart does provide a little bit more detailed guidance on how to measure for his pattern drafting instructions. It doesn’t prevent measurer / measuree errors, but I don’t see how that can be avoided entirely anyway.)
The other thing I wonder is whether the sewing community really reflects the RTW & general population. Maybe certain part of the population are already better catered for by RTW and therefore less interested in sewing? Perhaps there are more people on the extremes who’d be interested in taking up sewing just because the RTW don’t really offer them much choice / good fit?
LikeLike
This is fascinating and as a RTW 14 myself I get so frustrated when things aren’t in my size or I’m an XXXL. I have read though that while that may be the distribution of sizes, it doesn’t correlate to purchasing – those who’ve expanded their size range don’t find the curve of sales as you would expect – perhaps because women start buying fewer clothes? I don’t know. But it seems to be a no brainer to at least to to a US 16/18
LikeLike
When I was writing this I didn’t find much info on purchasing. But, I can think of a few reasons why an extended range in RTW might not sell (marketing, fit, style, etc.). For myself, after having 4 kids, I’ve found that I’m not just larger, but also differently shaped and it’s very difficult to get a good fit in RTW, even if my size is available. I also find few stores that carry styles I like, in my size. But I don’t think those reasons would apply to sewing patterns, where you get to adjust the fit and the style. And yes, I think it seems silly not to at least get to size 16/18.
LikeLike
Just a thought, but doesn’t this assume sewers are equally distributed over all sizes? I don’t have any data, and doubt it’ll be readily available, but I do think there’s a majority of sewers in a certain range (just based on the blogs I frequent).
LikeLike
I don’t think that’s true. There are a lot of reasons why bloggers might fall into a smaller range, then pattern buyers. 1) The internet is full of genuinely kind and thoughtful people (sewing bloggers, I’m looking at you!), but also quite horrible ones. And some of those people think it’s acceptable to make nasty comments about women’s sizes. It can be difficult to open yourself up to that sort of nasty attention. 2) Also, just because of societal pressure to be thin, some women who aren’t may not feel comfortable about their own size. Blogging takes confidence. 3) Also, if you are larger than average, you have a lot fewer things available to sew. 4) Larger bloggers don’t win sewing contests as often (if at all) and aren’t featured as often by sewing communities. There is just less publicity for them. 5) Maybe you just haven’t discovered their blog yet. Some suggestions:
The Curvy Sewing Collective
As Reasonable as Can Be Expected’s list of Plus Size Sewing Blogs.
En français, the group French Curves.
LikeLike
I think you are on target suspecting there are more plus sized sewists than are visible on the internet. I am larger than most available patterns, and I returned to sewing last year when I could not find a RTW dress to buy for my son’s graduation (my size was out there, but mostly in orange mermaidy shapes).
If you google plus sized sewing, there are a few great blogs out there, but there are also tons of people fat shaming photos and others talking about sewing as activism if you are plus sized and how “brave” the people are who put themselves out there. I just wanted to make myself some clothes, not be a brave activist.
LikeLike
I agree, no one should have to be “brave” to share an activity they enjoy. And yes, I think sewing offers more fashion choices for those who are outside the “regular” RTW size range. I know it does for me.
LikeLike
Great article. I wonder how the larger busted women who purchase patterns based on their upper chest size, vs their full bust measurement, fit into this study, if at all. It seems there are a lot of women who do this, which makes me wonder if there is a market for patterns based on a sloper using a c or d cup rather than a b cup. Of course, your study may, in fact, ultimately encompass this. I’m not sure. I read about so many women who must do a fba in order to get a pattern to fit, and not just use a larger-all-over size because the larger patterns are too big in the shoulder and neck area which are more difficult to alter. obviously this is is the reasoning behind the patterns that offer different cup sizes, which appear to be very popular but I don’t know how difficult, or costly, this would be to produce. Would it be more costly than offering a wider size range? Would it increase the percentage of the people who would buy the patterns?
LikeLike
I wonder this too. I started my pattern line with offering solely D-Cup sizing but I got a lot of requests for the other cup sizes or tutorials on how to make the pattern adjustments to smaller cup sizes. Now I offer A,B,C and D cup sizes for my patterns. I felt this might increase my customer base and if it does, the extra time I spend on developing this for my customers will be well worth it. Increasing my size range however, would be much more daunting. It is something that, after this discussion, is very much in the front of my mind.
LikeLike
GREAT post! It’s nice seeing what we, the not “normal” know in a perfect layout!
LikeLike
Just to add one more assumption (and it may have already been mentioned in the comments as I didn’t read all of them)…
You make a valid point about targeting a larger potential market in terms of sizing but that assumes that anyone of any size is equally likely to be a home-sewer. More likely I believe is that people at the extremes (i.e. not fitting into average sizing available on the high street) are more likely to take up sewing through the frustration of lack of available sizes/good fit for them.
LikeLike
I agree. It would seem to me that anyone outside the “regular”, below-average, ready-to-wear range would get more out of sewing. Or if they don’t sew, then they might be having someone else sew for them.
LikeLike
What a fantastic post!
As someone who is firmly within the size range that are covered by the pattern companies but larger than a model, I would like to add a thought. A lot of RTW companies use the excuse that their clothing isn’t designed for larger bodies so they don’t offer much larger sizes. I personally find it tiresome to see clothing in magazines, pinterest, etc and try to work out if the look – that is working so well for the very small model – would look good on my large busted size 10-12 (RTW) body. I understand it’s easier to make clothing look great on a model sized human, but I would really appreciate it if some designers applied their skill and creativity to figure out fun, cool, hip looks for people of larger sizes. And I don’t mean clothing designed to simply disguise ‘problem areas’.
LikeLike
I belong to the “do what you can to limit sizes”, club. Reason being, it is difficult to grade for a larger size range and do a good job (personally, I think it should be limited to 5 sizes, see http://www.fashion-incubator.com/archive/what-does-a-1-or-2-grade-mean/). The reason a lot of larger sizes don’t fit well is because they’re using the same base size instead of creating a separate (but identically styled) pattern for the larger sizes. In RTW, we manage this with size breaks. http://www.fashion-incubator.com/archive/what-is-a-size-break/
One way to reduce the number of sizes that one needs to produce is with niche styling that will only appeal to a given demography. While it is true that the average is X, ideally, the average should not be one’s customer. A pattern company can specifically limit a target range of sizes through style appeal. The reason the big 4 have problems is specifically because they are big -they’re forced to hit the broadest swathe of the market. In RTW, one way to manage larger sizes/differing styling, is to create another brand, an offshoot of the existing one. This isn’t something I’ve seen independents do but I think it is a worthwhile strategy to consider.
Lastly, this subject has been a long standing focus on my site (I published many posts similar to this over the years; this one specifically charts median distribution as do the “history of sizes” posts [3] http://www.fashion-incubator.com/archive/analyzing_sales_by_size/). Now, being that I’ve become a bore for dropping so many links that I risk getting filtered into the spam folder, I leave one last link to a post I published nearly 5 years ago, it’s called, “what if plus sizes made up 80% of the market?”. http://www.fashion-incubator.com/archive/what-if-plus-sizes-made-up-80-of-the-market/ Fwiw, the comments are educational. Point being, I think that if we’re not there yet, we will be soon so it’s time to strategize.
PS: Context: I have been making patterns in the RTW industry for over 33 years.
LikeLike
Hi! Great discussion! I thought I’d chime in with two cents on my process of grading bras, which I admit is crazy people business but I love it. After a certain cup size, which is somewhere in the middle of the size curve, the shapes don’t scale properly and sometimes become quite distorted. In my small experience, there are definitely size breaks that require new work, as Kathleen mentioned. So there is a need to create a new block with some slightly different style changes and grade rules. The same goes for the smallest end of the size range. If I were working in manufacturing, I’d probably choose to create two similar styles with distinct size ranges but each would need separate budgets. It’s fun stuff to think about for an online independent pattern business because we don’t have the same overhead costs. But we also want to do it well!
I am familiar with the SizeUSA study but have never seen it in depth. (Only the ATSM sizing.) Like Sharon, I am curious about the average bust size and how it relates to actual body size. Are they related in the study? The average bust size has grown but not always in relationship to the dress size. Over the last decade, the lingerie industry has exploded with brands doing full bust bras with very small bands. I love that Alexandra’s patterns are doing a D-cup–definitely fills a big niche (no pun intended!).
LikeLike