A $350B industry is in free fall
The fashion industry was already in serious economic peril before the virus even hit. In 2019, Barneys New York, Forever 21, Diesel, and many others filed for some form of bankruptcy protection. Since then, things have only gotten worse with additional notable brands succumbing recently including: Neiman Marcus, J. Crew, ALDO, and J.C. Penney. Many others are on the brink. There are several significant structural forces at play that are contributing to this cataclysmic decline:
- Working for home reduces the need for office fashion
- Significantly reduced dining and entertainment reduces the need for higher end fashion
- Loss of disposable income in the economy reduces spending across the board
- Curtailed interaction between the highly global workforce that contributes to fashion
- Lack of “business agility” in the typical fashion go-to-market process
Parallels Between Fashion and Traditional Software Development
It is this last point that I want to explore further. I have had the privilege of speaking with some fashion industry leaders recently to explore how “waterfall” thinking contributes to high risk and lack of business agility in the fashion industry and how lean and agile thinking might contribute to some of the solutions.
I reviewed some typical industry process flows recently and they often look something like this:
- Brand Strategy
- Fashion Forecasting
- Design Development
- Sample Development
- Production Planning
Remember “Windows 95”? That was the epoch of the big annual software release and all of the bugs, risk, and customer dissatisfaction that went along with it. The old software model was one where you decided on everything you were going to build up front, spend months designing it, more months building it (or having it built), etc. The result was very infrequent “drops”, dissatisfied customers, huge sunk costs, and lots of economic risk. If the drop didn’t go well, customers might not upgrade resulting in huge amounts of lost revenue.
When I look at the typical fashion process from end-to-end, I wonder if we have a similar issue. The typical fashion house probably goes through the whole merchandising process for the entire fall lineup all at once. If you try to design all of the clothes for the fall such as all of the slacks, all of the shirts, all of the jackets, all of the skirts, all of the dresses, all of the shoes, and all of the accessories, etc, then you would need many months to get through the whole process. You would also place your big bets and submit big orders for lots of inventory so that you can get the lowest cost-per-piece possible. The likely result would be that you are only able to come out with one or two big launches per year: the fall lineup and the spring lineup. That sounds a lot like the Windows 95 approach. This is a really risky way to develop new products because:
- You are locking into designs that might not succeed
- You are buying lots of inventory that you hope/pray will sell
- Your merchandise doesn’t get refreshed very often so there is no need for customers to come back to the store very often
- A competitor might come out with a hit and you have limited ability to respond
- A virus could hit and change everything
- The list goes on…
The truth is that this model doesn’t work very well and as proof, I offer the following. There is a huge multi-billion dollar segment of the fashion industry that is predicated on this model not working; they thrive on it. Ross, Marshall’s, TJ Max, Filene’s Basement, and many other discount retailers are counting on the traditional model to create a tremendous amount of overstock that they can buy up at pennies on the dollar. For example, at one point in 2019, H&M had approximately $4B of unsold clothes in inventory. The discount clothing market segment is approximately $100B in the US alone! The amount of waste in this system is extraordinary.
In a slower moving world and with enough fat margins, this model might work fine. But in a faster moving and highly dynamic world, it leaves you highly exposed to changes in customer sentiment, faster moving competitors, changes in the economy, etc. It also forces you to make huge bets … many of which do not pan out.
Fast forward 25 years. Nowadays, we develop lots of small software improvements and we deliver them continuously. This is how your phone apps are able to be updated almost continuously. An agile software approach might do something like this. Instead of designing all of the software at once, we decide on a few new features and focus almost exclusively on them. Instead of writing all of the software for the whole year, we only write a little bit of software for the the few features that we are launching right now. In software, we still design, develop, test, deliver, etc., but we only do a little at a time instead of trying to do it all at once. And we do this over and over again, every few weeks or months with the next small set of features. The result is that we are always delivering new features instead of having one big software update at the end of the year. Instead of making one or two huge bets a year, we make lots of small ones. The economic results of working this way have been revolutionary. Agile software development has been highly correlated with:
- Faster time to market
- Greatly improved responsiveness to market changes
- Lower cost
- Lower risk
- Improved customer and employee satisfaction
These steps that we go through to bring new software to market don’t really change that much. We still need to design, develop, etc. But the big game-changer is the change in “batch size” or the number of things that we work on at once. By only working on a few things at once, we move faster, lower our risk, lower our cost, and are able to pivot quickly when the situation changes.
What Would Agile Fashion Look Like?
An agile fashion model would first and foremost, lower the batch size considerable. We would not try to develop the entire spring lineup all at once. We would work much smaller. We might launch one ensemble right now … one dress, one skirt, one sweater, one pair of shoes, and one accessory. We would take the ensemble through the whole process and be able to come out with the new ensemble very quickly.
Then we do another ensemble in the same manner right afterwards so that we have a pipeline of drops that are coming out all the time. If we worked in this way, we would be coming out with new ensembles all the time. We could have a new drop every few months. This would allow us to test the market, respond to competitors, get feedback, and evolve our line throughout the year in a continuous fashion. We can see what customers are buying and not. We can see what competitors are doing and respond. Each new ensemble that we launch would have the benefit of the latest market information. Customers would have a reason to return frequently to the store or the website because the lineup is always changing.
My example may be rather extreme and perhaps the economies of scale are not able to support coming out with only one ensemble at a time, but perhaps we could do smaller groupings than we currently are and perhaps have 4 drops a year instead of two. Even this model would be moving at double the speed we are moving now and allow us to go to economic bets that are half of what we are making now.
Change is Hard But It Can Work
I know, all of this is easy to say but much harder to do. The economic order quantities would get all out of whack. We would pay a lot more per piece. Perhaps some of our key suppliers wouldn’t want to operate this way and we might need to find new suppliers, which is its own source of risk. Changes of this magnitude would be monumental in the industry; it is just not set up to operate this way. But it can operate this way and for some houses, it already does.
The Future is Already Here
As a wise person once said, “the future is already here. It just isn’t evenly distributed.” One of the largest fashion houses in world already operates in this way to some extent. Perhaps that’s how they came to be so successful. I’m talking about the Spanish fashion powerhouse, Zara. Zara is an economic giant in the industry, generating about $30B in revenue making them one of, if not the top fashion retailer by revenue. How do they do it? By bringing out many more collections per year than their competitors using a process that only takes 4 weeks front-to-back. Zara comes out with about 20 collections a year, compared to about 2 for almost everyone else! Admittedly, they are more vertically integrated than most houses in that they have the manufacturing capacity to build their own products. Their size certainly gives them some economies of scale too. But one has to wonder, how did they get to be so successful in the first place?
Roland Cuellar was the SVP of Business Agility at LitheSpeed.
Bring our experts to your team for an Agile assessment and jumpstart – contact us.