How to check the accuracy of graded patterns pt.3
Sorry for the delay in the continuation of this (by all accounts) most popular series ever (not) but as I mentioned, there was a bit of hacking in the interim. Hopefully by now no one is receiving an error message of a phishing attack when they load the site.
If you need to catch up, see parts one and two; today I’ll post three and four. Part three includes reasons why I don’t think you should sew sample size sets (and what to do instead) and part four is an email from a colleague who wishes his company did sew samples in all sizes. That entry is titled: Life in the trenches from the real world of a technical designer
The original question Mary asked was how to check a grade. Left for discussion was whether one should sew a sample in each size to test the grade. I don’t think you should. Mostly. I should qualify that.
Whether you should sew samples for each size depends on where you are in the gestation of your line. If you are new and perhaps tentative and uncertain that your sizing meets targets, you might want to sew samples of each size to test fit to your target customer. If you do this (and you should, but not routinely), pick one style that is representative of your basic body, aka whatever is closest to a block that you may have. The trick will be to find various bodies to try them on.
The danger of sewing full size sets and test fitting them can be endless iteration cycles. I’m not saying you should put out crap. I’m saying it is difficult to find one fit model that embodies your prototypical customer much less a whole size run of bodies that will. The problem is that you may end up using the fitting results of an outlying size body that is available to you, to adjust the other sizes (because grading cannot change shape, it only makes things bigger or smaller). Once you do that and re-cut another size set, the new pattern won’t fit the middle of the size run so you have to redo it again …see my point about endless iteration cycles?
Most people (even large firms, trust me on this) do not have access to a full size range of bodies to fit test their styles across all sizes. What people do instead (if they cut all sizes) is to measure the sewn samples and compare it to the grading chart to make sure it all works out. Maybe this sounds great to you. To me, it sounds like a sewing check, not a grading check. If you’re going to check grading, measure the pattern. Measuring sewn samples to make sure they match the pattern is not a grading check. I’m not saying you shouldn’t check the sewing output but you can’t go crying to the grader that the sizes aren’t right when the truth of it was that the sewing was off (see part four). If you’re using a full service provider it becomes more difficult to know anything. To prevent this from happening again, you may have to dig a little deeper and ask for nested patterns, the grade spec and the like. I realize this is a heady affair -when you’re small, using a full service contractor should simplify matters but it often complicates them when resolving problems. It’s a trade off. I respect your choices but it’s not one I would make personally.
A word of caution: Some people think that routinely sewing up a full size set to fit test on prototypical customers means one is more diligent, caring or dedicated than those who do not*. To me, sewing full size sets or even jump sets sends the message that one doesn’t understand how to prevent or manage non-conforming results. I realize that many people rely on partners (full service contractors) to do this part of the job so if this describes you, your controls must be different.
If your operation is large, outsourced and or very complex, resolution may be better suited to private consultation. In brief though, it should become an imperative that your firm has CAD software in house to check the graded pattern your vendor submits to you electronically (this may also mean hiring new skills in house but there you have it). You should require that the vendor submits a graded nest for analysis before any samples are made. Assuming you approve the grade after which you are sent samples, you need to measure the items to ensure these meet tolerances. If they don’t, you need to have a discussion with your vendor to ask why they sent you samples that didn’t meet the criteria you had established and they had agreed to. In my opinion, you are due a refund for non-conforming samples (and an apology for wasting your time). And sure, they usually say they’ll fix it the next go round but if they didn’t follow their own specs before making these samples, why would they do it in the future? It can make for a somewhat testy situation at times but I don’t see there is any other way to emphasize -from the outset- that adherence to spec is a condition of the contract. Hand in hand with this discussion is why you should pay for samples; it will cost less over the long run if you do.
But I digress. Here are other possible scenarios that would require the making of full sample sets to check the grade…~Oh wait. I should make a distinction here. There are two types of grade checks. One is to verify a grade you specified (or were recommended to use) is suited to the product and market, and the other type of check is to verify the grade didn’t negatively impact sewing; namely the joining of pieces. We can only discuss the latter. The former we’ve discussed a great deal (see Espionage for better sizing and part two as well as cruising the grading category). ~
Back to what I was saying, here are scenarios in which you might decide to selectively cut and sew a full range of sizes to verify a grade is suited to the product and market:
If you change your product styling in any substantive way.
If you change your fabrication in any substantive way (usually involves #1)
If you develop offshoot products for another customer profile.
Some examples of the above would be going from tailored silhouettes to flowing, loose dresses, going from wovens to knits (if you’d only done wovens or vice versa) or branching into toddler’s apparel when you’d been focusing on tweens.
*Some years ago, a home sewing personality cum “fit expert” enjoyed bragging that she had drafted her own pattern and sewed over 160 different fit samples for testing. Everyone in the room ooh-ed and aah-ed but I was appalled. It would be one thing to fine-tune and tweak a tire design, carburetor or solar cell 160 times but clothing? People’s bodies vary too much, even the same person from year to year and no pattern will fit everyone no matter how many samples you make. The result of too much testing is stalling in endless iteration cycles. She should have hired a professional to draft it. I would suggest that if you can’t get something to work within ten iteration cycles, there is a fundamental problem and you should probably move on to other styles that won’t be so problematic.