What we learned from 3 years of bra engineering, and what’s next

By December 3, 2018 No Comments

Three years ago, I quit my job as a software engineer and made the switch from engineer to bra-engineer and entrepreneur. I had a problem with bras, and I founded Bra Theory to build a better bra with an empathetic and mathematical approach.

Whenever someone asks me how Bra Theory is, I smile and say, “Good!” However, that’s the entrepreneur’s auto-reply in me speaking, since “I’m crushing it!” is the story we all want others to hear about our ventures.

The real answer?

Bra Theory is the most challenging, most fulfilling undertaking I’ve yet to experience.

Our goal is to deliver you a bra that fits your unique shape and size, saving you time and energy spent in the fitting room. In short, custom-made bras.

Our approach is to develop the Rx of the bra — the formula that maps between measurements on the body and the blueprint of the bra itself — and build you a 100% custom, algorithmically-informed bra. It’s the missing link in being able to scale a custom-made solution, which would make it more attainable than the existing alternatives.

Three years later, Bra Theory is not quite where we thought we would be.

It’s been a wild ride of highs, lows, successes, and failures. That being said, every failure has been an invaluable lesson: we are learning, through experimentation and doing the work, what it takes to get to the next level.

Without further ado, I’m excited to share the lessons that we’ve learned in the past 3 years about bra engineering and entrepreneurship.

Lesson #1: Tech founders? Don’t be afraid to ditch the high-tech solution.

In the beginning, I wrote a Python-based program that would take measurements as inputs and, after a few calculations, output the pattern (or blueprint) of the bra. We built it end to end, even including the seam allowances and labels that bra-makers require to sew the pattern.

From the archives — August 2016

From the archives — August 2016

It made sense to me: this was the Unique Value Proposition of Bra Theory. While the industry knows how to make a perfect fit through multiple fittings, the labor costs of the R&D of the bra patterns could only be offset by economies of scale. Traditionally, bra-makers can take up to 48 fittings to perfect the cut of a bra, and getting it right in 1 fitting is considered a “blessing.”

In order to make custom-made bras affordable, we had to reproduce that “blessing” of getting it right in one try.

Our approach was to parametrize the equation, writing the software to help minimize the number of fittings.

However, as we learned more about bra-making, it soon became clear that we were more often way off the mark than right. Rewriting code was taking way longer than we anticipated… and we began to feel the pain of having a clunky solution.

In 2016, we quickly learned a painful lesson: the important part was not the technology itself, but closing the feedback loop of measurements and bra-that-fits.

As much as I wanted to nerd out and build a sexy (pun-intended) 3D solution, we actually had to scrap our algorithm for a period of time so that we could move faster.

So we ditched the algorithm for something more agile, with fewer assumptions: we became a custom bra-maker.

We didn’t forget our vision. As our bra-makers taught me about the garment, I taught them about the algorithm. What we did had to be reproducible. Our resident bra-makers soon learned that whenever they “eyeballed” something, I was going to swoop in and ask a lot of annoying questions. We began to get feedback on the product, even if it wasn’t purely produced by code, in a way that could inform the code.

Though we pivoted from the 3D technology, it allowed us to simplify the problem.

In 2016, we redefined our goal: define the formula for a happy customer.

Old approach vs new approach (defining the formula for a happy customer)


Lesson #2: Be prepared to learn all the reasons why no one has solved your problem.

I started this journey uninformed and optimistic.


After preliminary research and speaking to industry experts, I was aware of the complexity of bra-making. I was even aware of the above graph.

However, I did not grok the time it would take to discover and understand the complicating factors.

Through more trial-and-error that I’d like to admit, we have learned through firsthand experience many of the variables behind bra-making. I will name a short list:

  • Whether or not a customer bloats
  • Dye color (black is more rigid)
  • Stretch properties of the fabrics, which differ from mill to mill
  • Temperature of the measuring tape at time of measuring
  • Elasticity/malleability of the breast, which in turn affects the shape of the breast when lifted and contained
  • The precision of how the underwire needs to be manipulated to sit properly below the breast root
  • The precision of pattern-making. Sometimes, millimeters can matter. Other times, you can cheat up to 1 cm of tolerance. When and where is key.
  • The positioning of the strap attachment at the back, which affects the inward tension that prevents the straps from slipping.
  • How fabrics behave when overlaid on top of each other, thus controlling stretch in different vectors

Is this an intimidating proposition? Occasionally. On other occasions, it’s fascinating.

Is this a solvable problem? Yes. Our process, for example, can include a custom underwire fitting that no other Ready-to-Wear brand can possibly include.

Different sized underwires

Different sized underwires

While we still have a lot of work to do, we haven’t discovered anything that says, “No, this is impossible.” Thus far, every failure has been an opportunity to handle a new exception in our algorithm.

Three years later, we’ve been through the trenches of informed pessimism, and are finally on the upswing to optimism. The key to getting past the trench?

Keep going. You need to maintain a light awareness of all the “no”s — but don’t focus on them — so that you can navigate a path through the challenges.

Lesson #3: Collaboration is key.

Through sheer stubbornness about the experimental method, Bra Theory was able to validate the hard and soft rules of bra-making.

However… did we have to discover it all first-hand, or could we have taken a bra-maker’s word for some of the novice mistakes that we made?

The answer is somewhere in between, which is why it’s important to collaborate with a domain expert as soon as possible.

Partnering with an expert means that someone can point out the pitfalls before we fall, face first, into them.

Exhibit A: the mysterious bra dent

Exhibit A: the mysterious “dent”

Notice the dent in the middle of the cup. While it is not too noticeable, especially in a photograph, we were certain that no customer would want a dent right where the fullest part of their breast should be. Our in-house team of myself and our freelancer designer were puzzled, until we consulted with an expert with specific industry experience.

“Have you measured the seams?” she asked. “I know it sounds silly, but the smallest difference can make a large difference.

A single millimeter — that was all that it took to knock us out for an afternoon. Since then, we’ve never made the same mistake twice, and measure more carefully… but it would be better to not make a mistake like this, even once!

Our key to success will be a beginner’s mind, an expert’s experience, and a healthy dialogue between the two.

The caveat to this lesson is that not every expert will understand your vision, and you’ll have to prove out your concept before they decide to all-in with you. I like to think of this as the modern “Non-Technical Founder Seeks Technical Founder” dilemma, except we’re talking about bra tech (#BraTech).

Lesson #4: Perfect is the enemy of good.

Like many others, I founded Bra Theory because I’ve always been on the elusive quest for the “perfect bra.” I even made a spreadsheet, years ago when I was still shopping on the market! For better and worse, that desire for perfectionism carried into the ethos of Bra Theory. We were ambitious, and aimed to deliver the Perfect Bra.

There’s a reason for that. From all the conversations we’ve had with women, we heard, again and again, how they feel like they’re “settling for the least bad bra.”

It’s an uncomfortable and often unsatisfying experience. Some women choose to delay shopping for bras for years until they absolutely have to; others have given up entirely on bras, disillusioned by the entire category.

What we're up against

Besides, we’re a custom bra maker… a custom-made bra better be perfect, especially if you’re paying $350 for it!

Unfortunately, the thought of “Perfect” slowed us down from “Good Enough”.

In the early days of Bra Theory, nothing was quite perfect enough to hit our “launch” button. I spent 2016 dog-fooding our bra, which meant that we worked through far too many prototypes for one person (which can lead to an overfitting error). By 2017, we had engineered one working bra: Bra Theory v2.1.1.17 for Mona Zhang, wide-and-tall-rooted 30E. In retrospect, a customer would probably kick us to the curb by fitting #7 — which, in retrospect, might have given us a wake-up call. If 17 fittings is unacceptable for a customer, it’s unacceptable for Bra Theory, too.

Our key lesson: it was never up to us to decide what’s Good Enough. It was always up to the customer.

We made bras for friends and family in November 2017 to validate “Good Enough,” and as soon as we got the green light, we hit the metaphorical launch button.

In 2018, we launched our Beta: it was time to invite strangers from the Internet into our studio, and do it live.

Lesson #5: Just say yes.

I had an assumption that in order to launch our Beta to paying customers in 2018, we needed a nicer space.

You see, we were measuring and fitting friends and family in a space that looked like my living room, because it WAS my living room.

“Are you sure you NEED a space?” asked my product manager, Brooke Kao, who is a strong advocate of prioritizing must-haves, not nice-to-haves. “You know, I went to an Etsy shop in a Manhattan mini-storage the other day… I thought I was going to get mugged, but when I got inside it was okay!”

I’ve heard the stories about Steve Jobs in his garage… but could we launch a bra company in a garage?

No, I thought, and I made a decision: I commissioned Kate Grau, a friend and fellow dreamer, to design a comfortable space on a budget.

Bra Theory SoHo studio

Bra Theory SoHo studio

Was it over-optimization? Yes. We realized, after asking our customers for feedback, that they didn’t need this space. They had expected to be in a WeWork, though they, too, wondered how we would finagle a nude fitting with “a single dry wall and seven sides of glass.”

Now we work out of The Wing SoHo, which is a perfect fit.

Feeling at home at The Wing

Feeling at home at The Wing

While we over-optimized this one, the important part is that we did what we thought we needed to do to move forward. We made the mistake of over-optimization. As a result, we stopped thinking about it as a limiting factor.

I agree that entrepreneurs need to learn how to say “no.” However, I’ve found it helpful to say “yes” when you’re just getting started, so long as it gets you closer to your customers.

Besides, there’s always a place in between “over-optimization” and “making your best educated guess.” When we explained the Manhattan Mini-Storage idea to one of our customers, she laughed and admitted, “Huh… yeah, I probably wouldn’t have come.”

(Alternative lesson: Do not launch a lingerie company in a Manhattan mini-storage.)

The other upside to saying yes is that you give yourself leverage to move forward. As soon as we committed to setting up our space, we did. As soon we said yes to a cadence of 2 customers a month, without the production capacity to guarantee it… we found the production capacity. If you put yourself between a rock and a hard place, and you’ll find your way through.

Lesson #5: Launch!

A mentor gave me advice three years ago: “Knowing you, launch earlier than later. It’ll never be perfect.”

I had resistance to this. I’m well aware that there’s a sense of fatigue and learned helplessness when it comes to bras, and I didn’t want to add to their pain by launching something that wasn’t perfect.

What they don’t tell you is that your early adopters will be the best people ever.

Your Minimal Viable Product will be far from perfect for your customers, and that’s the point: you’ll grow with them. Someone will take a chance on you, and when you find them, you will have found your tribe.

In November of 2017, we launched our website and Beta program. Our demand exceeds our production, which means we have to say no to those waiting patiently on our waitlist. We’ve been calling them in order to get to know them better.

Another point of resistance came up for me: how can we launch, when we can’t deliver? Will they want to give 15 minutes of their time, if we cannot solve their problem right now?

The answer is yes.

We have only learned, more and more, that the problem is real.

There is some disappointment, but above all, there is hope. This is what they say at the end of the day:


Lesson #6: Your health as an entrepreneur matters.

I have to admit — part of the reason that Bra Theory is not further along is because I had neglected my physical, mental, and emotional health for 25 years prior. It took a backseat to my goals, education, and career. As an over-achiever, Princeton English major, and auto-didactic software engineer, I had fancy titles, but little self-awareness.

Eventually, it caught up to me. After a bout of an un-diagnosed eating disorder from 2013–2015, acid reflux in 2016, and extreme IBS-C in 2017, I had to dedicate myself to troubleshooting my health.

I learned the hard way that if you’re an entrepreneur, most often, the limiting factor is you — which includes your health and happiness.

I don’t mean that in an unkindly way.

If you’re the one doing the brunt of the work to make the impossible possible, you need to treat yourself like your star employee. Unfortunately for anyone who has had an eating disorder, you deal with a lot of self-hate — and it is near impossible to run a business from that state.

The time I invested in myself, in turn, made me more capable of achieving my goals: I now have more energy, creativity in addition to analytical ability, and a growth mindset.

Lesson learned: working productively includes accounting for my feelings, desires, and personal life. Who knew — I’m human! So are you, and so are your employees.

Lesson #7: Remember who you’re building for.

Paradoxically, a custom bra is not for everyone.

The price is the main blocker for most people. The full custom-made bra (multiple fittings until you are happy with the fit) is $350. A trial of one measuring and one fitting is a $175 deposit, money back if you don’t like the fit. I am going to pull a Tesla and say that our ultimate goal is to use the cash from our premium product to invest in the R&D of an affordable ~$135 option, but it could be a while until then.

As of June 2018, our Cost of Goods Sold was $440 due to multiple fittings. We launched with the retail price of $550, with a strikeout: Name Your Price, a sales promotion for a limited time.

The experiment was ill-designed, and we discovered that people will name all sorts of prices when there is no price tag on a product. For example, we received several “$1” bids.

Then there was a woman — who we’ll call Cardamom — and she put down $550. She wrote a comment that I now keep on a post-it note:

“I have shed more tears than is healthy over this and the bra situation is the biggest sorrow of my life. I am hoping you can solve the issue and help women like me. I’ll be watching and following closely!”

Since then, I’ve noticed that Cardamom opens our monthly newsletter, every month, without fail.

Cardamom is out there, and there are more of her.

I started this venture because I had a problem with bras. But in a weird way, I’d wonder if I was alone. That’s what you end up believing — that you’re the odd one out, and that you’re the one with the “weird” boobs.

That’s why I am grateful for my breasts, and for this mission. Through Bra Theory, I have realized that I am not alone, and that there is something that I — and we — can do to change the world for the better.

Whenever I encounter hardships and challenges in Bra Theory (it’s entrepreneurship!), I think of Cardamom, and of myself.

I remember that, if I don’t succeed, there is going to be another little girl with breasts that are not your average 34B, with tears in her eyes from her experience with bras.

My biggest learning as an entrepreneur?

There are so many things that will go wrong.

But you only have to get a few things right.

We’re making bras, one at a time, and we haven’t given up.

The question, now, is…

How can we make more than that?

“At this rate, you’re going to become a custom bra-maker,” an advisor once warned me.

To scale custom-made bras, you must become a custom bra-maker.

To run, you must learn to walk.

To take off in your rocketship, you must build the rocketship.

Here’s the blueprint to our rocketship (2019):

  • 5 customers every month in NYC
  • Establish partnerships with local manufacturers to increase bras per month
  • 80% of customers are happy with the fit on on the first try, 20% by the 2nd try  (based on the Rx of your bra blueprint)
  • Launch the NYC experience
  • Nude
  • Colors
  • Strapless
  • The world

Thanks for reading.

P.S. Do you, or your loved ones, want a 100% custom-made, algorithmically-informed bra — and to help build this rocketship? Join our waitlist today.

What do you think of our lessons learned? Do you have any favorites, or counterpoints? Share in the comments below!

Appendix A: The Timeline


  • Conducted market research
  • Validated every hard and soft rule of bra-making known to womankind
  • R&D for the engineering of one bra


  • Launched the Bra Theory website
  • Conducted ~100 interviews to understand the problem
  • Launched the Alpha: friends, family, and referrals
  • Launched the Beta: several happy customers, and more customers in progress (30E –> 42KK)
  • Developed Algorithm v1.0, v2.0, and v3.0!
  • 17 fittings → 5–6 fittings → 3 fittings

Leave a Reply