Anyone who knows me knows that I’m not exactly a fashion-forward fellow. My friends constantly troll me about my shirts (straight out of 2004), my jeans (stained and 3-4 years old), my ripped jacket (yes, jacket. I own exactly one sport coat. It has never gone out of style–just ask me!). Perhaps they have a point. My wardrobe could use a little, uh, refreshing.
It’s not that I’m cheap. I’ve just never enjoyed the retail experience. This is true whether I am in a mall with my family, or whether I’m sitting at home in shorts and shopping online. I never seem to find what I want, and I don’t have the patience or stamina to search for hours for that perfect garment.
When I try to buy clothes at the mall, I am quickly overwhelmed by hundreds of items that I would never buy. I am clueless about “what’s hot this season.” I have no idea “what goes with what”. If I buy anything at all, I usually end up doing the Retail Walk of Shame, defeated, my arms clutching an outfit that would make Ron Burgundy proud.
Online shopping is no better. When I shop online, the recommendations that I receive make me feel that the retailers have absolutely no clue about who I am and what I like. The sites that I frequent will “recommend” on-Sale baby clothes or a Fall skirt.
My youngest is 15, and I’m a Suburban Dad, so, uh, no thanks.
Or this: I recently bought a vacuum cleaner on Amazon. Every time I visited the site for weeks afterward, Amazon recommended another vacuum cleaner. How many Kirbys does Jeff Bezos think I need?
This is frustrating because websites know a lot about me already. Theoretically, they should be able to use this treasure trove of data about me – the things that I have purchased in the past, how long it took me to make a decision before moving on to the next item, exactly which color combinations, patterns, and visual styles catch my eye — to serve up recommendations that are relevant to me.
Instead, they send recommendations based on what they think I may like, which are based on what they suppose other people like me like. This is a huge #Fail: I have unique style, color, pattern and shape preferences that are different from everyone else’s (my friends would say that my style just tilts more toward rips, stains, and the 1990s than your average person, but that’s another story).
So websites have all of this information about me. They could be using it to make it easier for me to find what I want. Yet they are trying to sell me stuff that they believe other people want. And that, when you think about it, is not smart at all.
The result is that I’m stuck in Ron Burgundy-land, which is a loooooong way from George Clooney-ville. Not where I want to be.
And yet, I am a VP Sales at an AI company that serves the Fashion industry. Mad Street Den sits right in the middle of the “Fashion AI revolution” that you might have been reading about, so there is hope for people like me.
I joined Mad Street Den for several reasons. AI is an exciting area. The founding team is made up of some of the brightest people I have ever had the pleasure of meeting. I can affect more change at a startup than at a bigger company.
But above all, MSD solves a real-world problem for me, which makes it a lot easier to get behind. Selling the Vue.ai product suite has been a fantastic experience. After all, feeling the pain and eating your own dog food is the best path to build a world class product that people want.
Using our computer vision AI, customers can find what they want and what goes with what. They do this in real-time, in an iterative way. And they do it well, because the recommendations that they receive are based on their personal preferences.
There are two components to how retailers can start providing better recommendations and choices to their shoppers.
The first is Computer Vision. Computer Vision systems “see” an image (like a dress in a product catalogue), and processes it like a human brain would. It sees the exact shades of its colors; whether it has ¾-length sleeves or is sleeveless; its hem length and neckline; whether it’s on a mannequin or a model; whether there are other people in the picture. And it remembers what this image looks like — just like my brain would.
The other part is Data Science. As I click different images on the website, the system learns the visual characteristics that I like. Maybe I’m looking for a dark brown leather jacket. I click on a Marc Brown mens’ jacket, and our system suggests visually similar alternatives — y’know, just in case the Marc Brown isn’t the exact one I want. When the website is powered by MSDs powerful technology, the site shows me relevant recommendations.
With every click I make, websites that are powered by Artificial Intelligence learn more about my preferences and make my experience better. For example:
- The next time I go to a Visual AI – powered website, my personalized home page will only show me the colors, patterns, lengths, shapes, and patterns that are relevant to me. No more searching for a leather jacket and the website suggests a pink onesie!
- When I purchase the perfect jacket, the system might suggest a nice-looking scarf that will go with it — and that I actually like.
- My confirmation email later that day might show me a pair of gloves that goes perfectly with the jacket — and that I would actually wear.
The result? It’s easier to find the clothes that I want that are unique to my true style. Heck, for someone like me, it’s easier for me to find a style that I like in the first place!
I save time because I’m finding the clothes that I need/want much quicker, without spending hours clicking around on different websites. Which I wouldn’t do in the first place.
Finally (and this is the best part), in my mind’s eye, I am more George Clooney than with Ron Burgundy, if only (and definitely only) with regards to the way I dress.