We Built A $40K/Month AI Tool That Speeds Up Clothing Manufacturing [Singapore]

Taime Koe

•

Published: September 15th, 2022

save

Taime Koe

Founder, Six Atomic

$40K

revenue/mo

Founders

Employees

Six Atomic

sixatomic.com

from Singapore

started April 2020

$40,000

revenue/mo

Founders

Employees

Discover what tools Taime recommends to grow your business!

social media tool-icon

Instagram

productivity tool-icon

Slack

Discover what books Taime recommends to grow your business!

📗

Want more updates on Six Atomic? Check out these stories:

Hello! Who are you and what business did you start?

My name is Taime Koe, the CEO of Six Atomic. We build automation technology to make real-time fashion the future. This means instead of buying clothes designed and made months in advance, you’ll instead buy clothes that have been made on-demand and personalized just for you.

We have three products: Synthesis, Inference and Envision. Synthesis allows users to grade any design to any size in seconds. It can also create 3D simulations of the garment, and output files ready for manufacturing. Inference is our body measurement validator and predictor.

This product calculates accurate body measurements from simple user inputs like their height and weight. Finally, we have Envision. It’s similar to Inference, however, instead of calculating body measurements, it recommends the best size for a consumer, and shows them how each size fits their body.

Our technology is used by businesses big and small, from enterprise clients like Perry Ellis, to small startups looking to take advantage of a more flexible, sustainable, and personalizable supply chain.

What's your backstory and how did you come up with the idea?

The apparel industry supply chain is great at certain things, like economies of scale, allowing us to buy clothes cheaply and plentifully. Alas, it is also really bad at certain things, like personalization, sustainability, and inventory risk.

If you need to design and produce months in advance, the likelihood of that product not selling as expected when it finally gets to market is all but guaranteed. In fact, for every retail dollar spent on clothing, about 70 cents is lost due to inventory being sold at discount, or thrown out entirely. This unsustainable hurt profitability and prevents consumers from accessing any sort of personalization in their clothing options.

There’s an obvious solution to this - make it on-demand (or in real-time). This means you only produce what has been sold already. If you can do this, you can personalize the design and fit of each product to each consumer. Brands can pay their suppliers more while still profiting more, because they’re not losing 70% to inventory risk, leading to the possibility of reshoring manufacturing. Everyone gets a better deal overall - brands, consumers, suppliers, and the environment.

The apparel supply chain is however very complex and requires many experts and vendors involved in the process from design to production. Therefore, to make on-demand production scalable, we need a lot of automation. That’s where Six Atomic comes in - we build automation technology to make real-time fashion the future.

Take us through the process of designing your MVP

We first focused on the most critical part of the apparel supply chain - the pattern. The pattern is essentially the blueprint for every item of clothing. It’s the geometry that is used to cut out fabric and sew it together into an item of clothing. By automating the creation and manipulation of patterns, we’re able to make any garment fit anybody. We’re able to export patterns to the right format so any factory in the world can sew them together. We’re even able to create photorealistic samples virtually.

We spent many years researching pattern-making theory, from Victorian-era drafting texts to modern Japanese patterning methods. At the end of this process, we sequenced the ontology of patterns and built algorithms to create and manipulate them based on the inputs provided. So if I took some measurements off my body, I could input those measurements into the algorithm and generate a unique pattern just for me.

This technology has been applied to solve several different problems. This includes generating custom patterns for individual bodies at scale and building modular design libraries where brands and consumers can mix and match components to create new and unique designs. Of course, this is just the starting point of what’s possible, with the ultimate goal of creating a real-time supply chain by connecting the important dots from design through to production. We’re working closely with our enterprise clients to build out the end-to-end capability, so don’t be surprised if your favorite brands start offering personalizable products in the near future.

Describe the process of launching the business.

My partner and I invested some of our money upfronts to get started, and we’ve been bootstrapping ever since, primarily through revenue from enterprise client projects. Bootstrapping has been good for our business, albeit difficult at times, because it allows us to be the sole drivers of the direction of the business, and invest in areas that we believe are important for the future.

Bootstrap as long as you can. To be in control of your vision with modest means to pursue it is better than a billion in the bank with strings attached.

We’re in the process of increasing our recurring revenue through our SaaS offerings. The products I mentioned earlier are all turn-key solutions, so startups can quickly and easily sign up and start benefiting from our technology. Enterprise clients continue to be an important area to learn and build out our end-to-end solution, and those capabilities will eventually trickle down into our turn-key offerings.

We focused on enterprise clients first because there was so much we could learn from them, as well as apply our technology at scale. The first key enterprise clients came through speaking at conferences, referrals from other people in the industry, and generally hustling our way in front of people who could benefit from our capabilities. We learned a lot about how products are developed, and how our technology must fit into that flow to be successfully adopted. We learned about building robust application architectures to ensure they’re enterprise-grade. We also learned about how our technology roadmap could align with the strategic objectives of the world’s biggest apparel businesses.

From cutting our teeth at the very top of the industry, we’ve been able to craft our turn-key SaaS offerings. Of course, these products need to be refined and shaped based on user feedback, and that presents an interesting challenge for us, however, we have much more confidence in product-market-fit given our enterprise experience.

Since launch, what has worked to attract and retain customers?

For enterprise clients, it has been presenting at conferences, publishing op-eds in industry publications, and referrals. The main objective is to establish ourselves as domain experts in our field. Only then are they willing to trust us to listen to their needs and propose solutions to their problems? We have a particular process when we start working with a new client.

The first step is to put together a proof of concept so they can tangibly see the solution working. From there we’ll complete a scoping study, speaking to their stakeholders to fully map out the problem and opportunity. Thereafter there’s usually an R&D stage where we build specific capability for the client. Development then follows, and we wrap up with deployment, maintenance, and ongoing support.

For SaaS offerings, we have been using a few different approaches in our funnel, including email marketing, SEM, and SEO. Email marketing includes cold emailing with several automated follow-ups. Our email content is quite personalized, short, and to the point. We want the emails to come off as sincere and client-focused, rather than shoving our solutions down their throat and hoping for the best. It’s somewhat effective, but a cold email is never as convincing as a warm introduction.

For SEM and SEO, we try to identify search keywords and phrases that align with our product offering. The challenge we face is low search volumes given our offerings are rather a niche. That’s not a huge problem in the B2B space because we get scale through quality rather than quantity. All in all, our traffic continues to grow and our CPAs are coming down month-to-month, so we’re heading in the right direction.

How are you doing today and what does the future look like?

We believe we’re in a formidable place. We have committed enterprise clients that share our vision for a real-time supply-chain, and want to realize it as much as we do. We have solid SaaS offerings that solve critical problems from startups and SMEs that are affordable and easy to use.

Our revenue continues to grow as we continue to bootstrap. These are all great fundamentals for any young company, and it’s a credit to our team and the support we’ve had to get to this point.

I think you’ll start to see excellent showcases of our vision through our enterprise clients as they roll out their various initiatives. This will inspire further investment and growth in the real-time supply chain, and we’re in a great position to capitalize on this.

Enterprise income will continue to be the bulk of our revenue for a few more years, but with our SaaS offerings, that should begin to change as smaller companies put our technology at the core of their business.

How this all translates to numbers is perhaps counterintuitive. We’re an automation company, so growing our team in step with our business growth would mean we’re doing something wrong. We like to work smart and not hard, preferring to build algorithms to do much of the heavy lifting as our company grows.

From a revenue perspective, we know that the cut-and-sew apparel industry today is worth ~$500 billion. Ushering in a revolutionary supply chain should be worthy of a few percent of that pie.

Through starting the business, have you learned anything particularly helpful or advantageous?

Firstly, listen to your clients and never assume you have the right solution for them. When we do our scoping studies now we completely forget all the tools and capabilities we have built and just listen. We may learn that nothing we have to offer is what they need, saving both the client and us wasted time and money. It might sound weird to turn down potential revenue, but integrity and focusing one’s attention are more valuable.

Secondly, hiring the right people is key to business success in the face of growth. A good hire will be an extension of yourself, but hopefully much more competent at the role than you are. If you hire people that do a worse job than you can, you’ll end up micromanaging them and struggle to grow.

Thirdly, bootstrap as long as you can. I have heard so many founders lose control of their vision because they raised capital. Investors want to see results in certain timeframes, and when you’re in that early stage of finding product-market-fit, testing various hypotheses, and seeing what sticks, the last thing you want is someone breathing down your neck. To be in control of your vision with modest means to pursue it is better than a billion in the bank with strings attached.

What platform/tools do you use for your business?

Our team uses GitLab for development and project management. Sprints allow our team to prioritize and react to issues fast, and collaborate on finding solutions or providing support. As for communication, we use Gitlab for asymmetric communications, Slack for symmetric communication, Miro for visual collaboration and planning, and Google Workspace for email, calendar, etc. For designing, we mainly use Sketch.

What have been the most influential books, podcasts, or other resources?

One of the most influential books I’ve read (and get my team to read), is Crossing the Chasm by Geoffrey A Moore. It talks about bringing new technology to the market and how to cross the “chasm” from early adopters to mass-market customers.

Because we’re a technology pioneer, it’s difficult for people to understand how our product can solve their problems. We cannot approach it the same as products with a lot of precedence in the market. So if you’re bringing a new solution to market, this is a must-read.

Advice for other entrepreneurs who want to get started or are just starting out?

Get an MVP out to market and test with real users as soon as possible. By MVP I mean a product that is bug-free and offers value, albeit heavily paired back in the interests of getting feedback from prospective customers asap. You want your customers to shape your product rather than build it in isolation. I see too many people using models, canvases, and other worksheets as an alternative to going out and building. Or on the flip side, developing the perfect app which they eventually learn no one wants.

Also, just like you build an MVP, you should build your customer acquisition funnel. This means launching your campaigns as early as possible so you can assess the demand and market size for your solution. There’s nothing worse than building a tool that you struggle to sell. There should be as much attention put into customer acquisition at the earliest stages as there is on the product itself.

Are you looking to hire for certain positions right now?

We’re not currently advertising any positions, however, we’re always eager to connect with skilled algorithm engineers (machine learning or mathematical). You can reach us by email at [email protected]. We’re always open to talking to skilled talent worldwide.

Where can we go to learn more?

If you have any questions or comments, drop a comment below!