This blog post is sponsored by Bluezone, the independent show for denim and sportswear by Munich Fabric Start. Register for the next show here!

Call me old-fashioned but—like many of my fellow denimheads—I have a penchant for ‘old stuff.’ I like my jeans loomstate; my boots Goodyear welted; my wristwatch vintage.

At the same time, I love the convenience of modern technology. In fact, I don’t think I could live without it!

Just think about how you would talk to your friends, do your email, catch up on the news or listen to music before 2007; before the iPhone. Plus, it tells time just as well as (or maybe even better than) my 1944 Omega ‘Suverän.’

Digitalisation is changing everything, including the fashion industry. And the new technologies can prove heaven-sent:

“The fashion industry is often gridlocked in a paradox that’s rooted in the tension between its creative side and the bottom line,” Elizabeth Doupnik, WDD’s associate editor, wrote in a blog post earlier this year. “And in a retail environment undergoing extreme upheaval due to the demands of consumer centricity, this tension is being further strained.”

The denim business is on the verge of a digital revolution. It will be disruptive, but it will also create a lot of new opportunities as the digital becomes more and more intelligent.

Disclaimer: This post was written and published in December 2017. AI has since then evolved quite a bit and continues to do.

The Contrast Between Man-Made and Machine-Made Creativity at Bluezone

Building on the ‘denim beyond seasons’ theme, the Bluezone show in January 2018 was titled MAN+MACHINE. “Our goal is to communicate disruptive changes in our industry,” the show’s denim curator, Panos Sofianos tells me.

The show will address how software-generated creative processes and analysis of sales data can help brands and makers create better products.

It focuses on how creativity driven by humans can be combined with design and range-construction that is generated by software and ‘big data’,” says Bluezone’s creative consultant, Tilmann Wröbel.

It’s about the contrast between human (illogical) creation and machine-made (logical) creation, and how humans can improve using this technology.

It’s not ‘man versus machine;’ it’s ‘man and machine,’ Bluezone’s art director, Jo Baumgartner, elaborates. “It’s about an impetus to human creativity, whether handcrafted or digitally programmed.”

“We believe this presents enormous potential for denim-driven creatives as much as for denim-related algorithms,” Tilmann reasons.

In this sponsored blog post, I discuss this topic with a handful of influencers from our industry.

How the Denim Industry Already Uses ‘Big Data’

The best way to find out whether what you’re making is any good is to look at how it’s selling. Consumers express their opinions and preferences through what they buy, Lenzing’s Tricia Carey so rightly argues.

Like most suppliers, brands and retailers, the fibre maker studies reports from sales data closely to understand consumer preferences and adapt their product development and marketing message accordingly. They do the same in laundries:

When I was at Martelli, we checked sales data for washes,” says Giovanni Petrin, “and we analysed the most popular ones to understand trends and help our clients prepare collections,” the former manager of Martelli, who’s now consulting together with his son, adds.

These days, it’s become a lot easier to process vast amounts of data about trends, consumer preferences, and what’s actually selling (and what isn’t). Working in a scientific way with sales analysis is crucial to help avoid mistakes, designer Piero Turk argues.

In fact, it’s indispensable for those who mass-produce, says Alice Tonello. The Italian manufacturer of denim laundry solutions use sales analyses as input for their work; to have a direction and to understand what path to follow (or avoid).

What is ‘big data’?

The term ‘big data’ describes any voluminous amount of data that has the potential to be mined for information. “It helps human see multiple options to accelerate the decision-making process,” says Artistic Milliners’ Ebru Ozaydin.

If you want to learn more about how big data can help not only the denim industry but also revolutionise medicine, and even potentially alter our understanding of the universe, do yourself a favour and listen to this episode of NPR’s TED Radio Hour show.

Designer Stefano Aldighieri points out that “a proper analysis of past performance is always handy when you start creating your next collection.” The only caveat, he adds, is that “as good as this analysis can be, it’s only a picture of past performance; it does not tell you much about the future.”

Ebru is on the same page. “Sales analysis is a ‘post-activity’ where you analyse what you have already performed,” she tells me.

Sometimes, an article might not sell at first,” Piero suggests. “When that’s the case, you have to take a decision. If you believe in the item, you might keep it in the collection, but if you look at numbers alone then you’ll take it out.”

“The ideal is when you combine big data sales analyses with the out-of-the-box creativity of human minds,” Alice concludes.

Digitalisation Introduces Cognitive Technologies

In the near future, though, we will also see the next step of digitalisation in the denim industry as cognitive technologies such as AI (artificial intelligence) starts designing products based on big data.

AI has been the villain in countless science fiction books, movies and TV shows. But while real-life AI is moving fast, it isn’t as scary as HAL 9000 from Space Odyssey; more like R2-D2 and C-3PO.

In essence, AI is about teaching a machine or a piece of software how to identify nuanced and complicated things with human-like awareness, and then react accordingly,” argues Katie Smith from EDITED.

AI gets beneath all the data we’re collecting to drive insight. It’s currently capable of spotting trends, predicting consumer preferences, and remixing existing designs to create new styles.

What is AI?

Before you can have AI you need machine learning, which is where computers learn from their actions and mistakes and continuously improve.

To elevate machine learning to AI, you need deep learning, which focuses on understanding features and representations within data to generalise better to add extra dimensions to algorithms.

An algorithm is like your friend who’s really into vintage Levi’s jeans; he’s great at something very specific. Deep learning is an all-rounder; it’s the guy who knows everything about the fashion industry, and who seems to always intuitively understand new things with minimal effort.

Compared to algorithms, deep learning can spot patterns and subtle features, just like the human brain.

A well-known example of deep learning is IBM’s cognitive computing platform, Watson. It first garnered attention in 2011 when it beat two Jeopardy champions. Before Watson, IBM built DeepBlue, the computer that defeated world chess champion Gerry Kasparov in 1997.

In the world of fashion, Watson has also been used to analyse trends and correctly predict the colours and patterns that dominate consumers’ closets. Democratising trend analysis like this will have a profound impact on our business.

And with today’s deep learning, all the data we’re collecting becomes incredibly insightful, and it allows software to automate decision-making processes to help us humans do our jobs better and faster.

The Advantages and Opportunities of Cognitive Technologies

With constantly and ever-faster changing trends, the ability to apply real-time and even predictive insights will be a major advantage for the denim business.

The human minds that envision trends can be enhanced by the analytical capabilities of machines, IBM’s VP and GM, Steve Laughlin, says in a Quartz article. It’s a tool to inspire designers and minimise risk.

Cognitive technologies such as AI don’t replace the creative process; they enhance and accelerate it. “Modern technology is unleashing creativity—not stifling it,” Laughlin argues.

According to the article, the global apparel market is worth $3 trillion, which is 2% of the world’s GDP. Yet it’s a market that essentially relies on a lot of guesswork.

That’s why the implementation of technological advancements such as deep learning and AI are top priorities for the big players.

“Being able to predict the ebbs and flows of consumers’ whims can mean multimillion-dollar differences in annual earnings,” Laughlin argues.

The Artificially Intelligent Designer

Amazon, one of the new players in fashion, is betting big on AI. The online mega-retailer has developed an algorithm that can design clothing by analysing images and then applying what it learns to generate new items from scratch, MIT reported back in August.

The algorithm was developed by Amazon’s Lab126 using a tool called generative adversarial network (GAN), a new technique in AI. It’s able to learn from raw data, in this case images, and then design clothes.

Imagine how this could be used to harvest input from sources like social media to help designers come up with new creations.

Amazon’s artificially intelligent fashion designer raises questions about our industry’s creative future, says tech expert, Tak Lo. He argues that creativity is less complicated than we think and that it’s actually quite standardised.

Designers essentially analyse previous work, assess what’s trending, and then uses this data to create new styles, he argues. “An algorithm can follow the same process and create fashion,” he provocatively points out.

Piero Turk, Ebru Ozaydin, Alice Tonello, Tricia Carey, Stefano Aldighieri and Giovanni Petrin
From left: Piero Turk, Ebru Ozaydin, Alice Tonello, Tricia Carey, Stefano Aldighieri and Giovanni Petrin.

Can AI Be Creative?

At this point, (human) designers reading this might be thinking “a computer can’t do what I do!” And despite AI’s advances, there’s still widespread doubt over whether it can ever become truly creative.

The acclaimed expert on digital culture, Gojko Adzic, believes that AI in fashion design will mainly affect the fast fashion and high-volume sectors of the market, where, as he says “there’s not much genuine innovative design, anyway.”

Personally, I also believe we’ll continue to need human designers for the foreseeable future.

The influencers I’ve talked to agree and argue that the key difference between a logic creation made based on analyses and a creation made by a creative human comes down to two things:

Intuition and Creative Sparring

“Creatives are inspired by their environment, people on the street, influencers, and art,” Tricia Carey argues. Data, in her opinion, is there to guide and give some insight into demographics and behaviour of the past.

We cannot use data as a crutch as we need to take risks. Creativity will drive the industry forward and sometimes you have to go with your gut!”

Piero Turk is a great example of this. “I’m not an expert in working with big data,” he tells me. “I use my gut feeling.”

Data collected and mined by computational processes can certainly help to discover patterns in large datasets thanks to artificial intelligence and machine learning, Ebru Ozaydin argues.

However, intuition still has an important role to play to make decisions,” she adds. “Machines still do not have ‘gut feeling’ to trust as decision-making process is not purely logical but also emotional.”

There’s no interaction with a machine, Stefano Aldighieri points out. “It’s impossible to bounce off ideas and generate something that may look crazy which could be the next ‘big thing’.”

Thinking ‘outside the box’ is something the box cannot do!,” the US-based Italian designer adds.

It’s difficult to imagine creativity closely linked to numbers, Alice Tonello speculates. Still, she acknowledges that in-depth analysis of a specific product can help you develop the right creative strategy and that it probably makes the new ideas more effective.

Software is increasingly central in today’s work, but creativity is still an unsurpassed quality,” she says, “because it’s based on know-how, knowledge and human heritage. Something that has to be ‘felt’ and not just read.”

Sales analysis is indeed an extremely valuable and powerful tool that can help makers, brands and retailers weed out bad performers and identify potential winners and underlying trends, Stefano recognises. Still, he believes the creative aspect humans will have the edge, at least for the foreseeable future.

In the end, creativity is one of the hardest things to automate.

Supported by Munich Fabric Start’s Bluezone

This blog post is sponsored by Bluezone; Munich Fabric Start’s independent trade show for the denim and sportswear community.

The family-run Munich Fabric Start was established in 1996. Twice yearly, it attracts 20,000 fashion professionals to Munich. Bluezone was launched as the first denim-dedicated show in 2003. Today, it hosts more than 100 carefully curated exhibitors.

The show caters to all your sourcing needs: well-established ‘all-star’ mills; ‘catalyzers’ that create future trends in denim; and how new technologies and laundry solutions can make denim more sustainable.

Register for the next show here!

Disclaimer: Denimhunters was invited to Munich Fabric Start by the organisers. However, none of the individuals or companies mentioned in this blog post are affiliates.

Author

Thomas founded Denimhunters in 2011 and built it into a voice for the denim industry and community. Today, he helps companies grow and gain market shares using his STORY/SELLING BLUEPRINT.

1 Comment

  1. Martin Schaefer Reply

    sales figures and expected margins bring designers back to reality.

Write A Comment