What problem in the fashion industry made you decide to build Clothink?
I kept seeing the same problem from both sides of the product development process.
Through my manufacturing business, we were regularly approached by startup founders with strong product ideas but very little understanding of how to turn them into something a factory could accurately quote, sample and manufacture.
At the same time, we were managing that journey for clients ourselves, taking an early idea through design, visualisation and technical development before it could move into sampling. Each stage involved a lot of manual work, with information often having to be revised or recreated as the product progressed.
We initially began exploring whether AI could make that process faster and more connected for our own team. But it quickly became clear that the wider problem was the gap between having an idea and having a properly developed product that a manufacturer could work from.
We built Clothink to bring those stages together and give founders and product teams a clearer, faster route from an initial concept to a product that is ready to develop with a factory.
There are already countless AI image generators. Why did you focus on product development and production rather than just design?
Generating an attractive fashion image is relatively easy now. The difficult part is turning that image into a real product.
A brand still needs to decide what the garment is made from, how it is constructed, how it should fit, what details it includes and how it sits alongside the rest of the collection. All of that then needs to be communicated clearly to a supplier. That’s the gap Clothink is focused on.
We aren’t trying to help people generate endless images that never go anywhere. We want to help them take a promising idea and develop it into a product that has been properly thought through, clearly specified and is ready to move into sampling with a manufacturer.
What are the biggest mistakes fashion startups make when trying to turn an idea into a physical product?
The biggest mistake is assuming that a factory will develop the product for them.
We regularly see founders approach manufacturers with little more than an image or a basic sketch and expect an accurate quotation and a first sample that reflects what they have in mind. But a factory still needs to know the fabric, construction, measurements, trims, finishing and intended quality level. If those decisions haven’t been made, the supplier either has to guess or ask a long list of questions before anything can move forward.
That’s where inaccurate pricing and disappointing samples often begin. The factory may produce something that technically matches the information provided, but not the product the founder had imagined.
Another common mistake is trying to launch a large collection immediately. Every additional style creates more development work, sampling costs, minimum order requirements and opportunities for something to go wrong. A smaller, well-developed range is usually a much stronger place to start.
Many founders underestimate just how many decisions sit between having an idea and having a product that is ready to manufacture. The more clearly those decisions are made at the beginning, the smoother and less expensive the development process tends to be.
Many founders think AI will replace creative jobs. Do you see AI as replacing designers, or helping them work differently?
I see it as helping designers work differently.
Good design isn’t simply about producing an image. It involves judgement, taste, commercial awareness, technical understanding and the ability to make hundreds of small decisions throughout the development process.
AI can help someone explore ideas faster, test different directions and remove some of the repetitive work, but it still needs a person to guide it and decide what is actually right for the product or brand.
For experienced designers, I think it can become another tool in the creative process. For founders without a formal design background, it can make parts of that process much more accessible.
The strongest results still come from human direction. AI can speed up the route, but the judgement about where you are going still belongs to the designer.
Fashion is often criticised for being wasteful. Could AI help brands reduce wasted samples, materials and production costs?
Yes, particularly during the development stage, where a surprising amount of waste is caused by unclear decisions rather than the product itself.
Samples are often made simply because a colour, detail or proportion was not properly resolved beforehand. The sample arrives, the brand decides the pocket is too large or the artwork is in the wrong position, and another one has to be made. Those changes may sound small, but each round uses more fabric, trims, freight and development time.
If brands can explore more of those decisions digitally, visualise the product more accurately and give the factory a clearer brief, there should be fewer avoidable samples and corrections.
AI isn’t going to solve overproduction or make the industry sustainable by itself. But it can help brands make more informed decisions before materials are cut, which is a practical place to start.
One concern around AI-generated fashion is that it can produce ideas that look good on screen but don’t work in real life. How do you tackle that challenge?
That’s one of the biggest challenges with AI-generated fashion at the moment.
An image can look convincing while showing fabric behaving unrealistically, construction that would be difficult to achieve or details that would be far too expensive to reproduce. Coming from a manufacturing background, we know that a visual is only useful if the design can eventually be translated into a physical product.
That’s why Clothink doesn’t stop at image generation. It helps the user define the product in more practical terms, including its materials, construction, fit, trims, measurements and technical details, so the idea can be developed into something a supplier can actually work from.
AI still can’t replace technical review, fabric selection or physical sampling, and we don’t pretend that it can. The aim is to help brands enter those stages with a more clearly defined product, rather than treating an attractive generated image as a finished design.
What’s been the reaction from designers? Are they excited about AI or still sceptical?
From the designers and product teams I have spoken to, the reaction is mixed, which I think is healthy.
Some immediately see the value in being able to explore ideas, colourways and collection directions more quickly. Others are understandably cautious, particularly when so much of the conversation around AI has focused on replacing designers or generating large volumes of generic imagery.
I think that scepticism starts to change when people see AI being used as part of a genuine development process rather than as a novelty. The interest tends to come from the practical applications: getting to a first concept faster, testing alternative directions, creating product visuals or reducing the time spent assembling the first version of a tech pack.
Designers should question these tools. They need to earn a place in the workflow by saving time or improving the outcome, not simply by producing something that looks impressive.
How do you think AI will change the way fashion brands develop products over the next five years?
I think the biggest change will be how quickly brands can move from an initial idea to something they can properly evaluate.
At the moment, a lot of time is spent developing and agreeing the first concept, colour direction, product visual, specification or range plan. AI can shorten that stage, giving designers and product teams more time to refine the idea and develop the physical product properly.
I’ve seen brands spend so long agreeing the creative direction that, by the time it is signed off, there is very little time left for sampling, testing and corrections before orders need to be placed for the next season. Physical development then becomes rushed because too much of the critical path has already been used.
AI should also make development more iterative. Brands will be able to explore different colours, fabrics, details and range options much earlier, before becoming too committed to one direction.
For smaller brands, it should also make some capabilities that previously required a much larger team more accessible. They’ll still need designers, garment technologists and manufacturers, but they should be able to enter those conversations with a clearer and more developed product.
The brands that use AI well won’t necessarily be the ones generating the most ideas. They will be the ones using it to make better decisions sooner and leave more time for sampling, testing and getting the physical product right.
What’s something people outside the fashion industry don’t understand about how difficult it is to get from a sketch to a finished garment?
People often assume that once the design looks right, most of the work has been done. In reality, that’s often where the work begins.
A jacket, for example, can involve decisions about the outer fabric, lining, padding, zips, pockets, seam construction, stitching, labels, measurements, grading, testing and packaging. Each decision affects the appearance, performance, price and manufacturability of the product.
Even a small detail can create problems. A pocket may look right in a drawing but be difficult to sew, badly positioned in larger sizes or too expensive for the target price.
A factory can’t manufacture an idea. It needs clear instructions and dozens of interconnected decisions. Translating a creative concept into that level of detail is one of the hardest and most underestimated parts of developing a garment.
If you could solve one problem for fashion founders and independent brands, what would it be?
I would make product development less of a black box for people coming into fashion without an industry background.
A lot of founders have a strong idea and are willing to put the work in, but they don’t know what good looks like at each stage. They may not know whether the design is detailed enough, what information is still missing, why a sample hasn’t matched their expectations or when the product is ready to take forward with a manufacturer.
That uncertainty can become expensive very quickly. Founders can spend money developing products that were never clearly defined or repeatedly revise samples because important decisions weren’t made early enough.
I would like Clothink to give them a clearer route through that process, helping them develop the idea, understand the decisions they need to make and prepare better information for the designers, garment technologists and manufacturers they work with.
The goal isn’t to turn every founder into a technical expert. It’s to stop a lack of industry knowledge from becoming the reason a strong idea never reaches production.
