Empower Your Business With The Magic Of Machine Learning!
- +30 210 9701777
- Aristotelous 83, Athens 163 43
- contact@mobiplus.co
Empower Your Business With The Magic Of Machine Learning!
Stitch went public on the NASDAQ exchange in 2017.
The company was launched in 2011 as an online subscription service of personalized clothing shopping for Women,Men and Children.
In 2018 it had a revenue of $1.25 Billion.
It uses the Choice Architecture philosophy so that they have a business whose role is to advise clients on how to make their appearance better.
This is the central Architecture of the enterprise.
The No. 2 in the business after CEO Katrina Lake is Eric Colson Chief Algorithm Officer -CAO who was the VP Data Science & Engineering at Netflix.
CEO Katrina Lake says:
Data science is our culture.
The heart of our business.
We build our business algorithms around our customers and their needs.
The Data Science department reports directly to me.
The first thing they ask the customer when they enter the service is detailed information about personal preferences, numbers and money they want to spend.
They use a kind of game where the customer sees mixed clothes and accessories and by swiping left or right they can say if they like what they see.
This gives the Recommender important information about customers basic characteristics.
Recommender selects the best recommendations from about 700 Brands that the company has and several others that are brands of the company.
Subsequently these recommendations go to one of the 3500 Stylists that the company has for review.
The stylist selects 5 products — fix-and send them to the customers every 15 days, month or quarter depending on the customer.
Included in the package is personal advice for the combination of the items.
After receiving the package, customers review each product in their personal account on the site.
This way the Recommender can and subsequently learn the needs of the customer and evolve.
Customers provide rich data during the review such as :
Fit , How does it looks on them , whether they like the style , and their other thoughts , such as …Fits nicely on the body but is narrow in the shoulders….
This detail is incredibly useful in the Recommendation Engine.
To learn how the top companies in the world increase revenue up to 30% from existing customers using Recommendation Systems and Personalization read the book below.
https://mobiplus.co/ebooks/customer-rediction/
The second thing they ask customers to do is create a personal clothing and style album on Pinterest.
Choosing clothes and accessories from millions of recommendations of Interest they create their personal style.
They then use Machine Vectorisation on the above images and extract features from the images that are useful for Recommendation Engineering.
They then use Convolution Neural Network architectures and in particular Alexnet , to match the attributes of customer responses from the album in Pinterest with their own products.
This Neural Network matches the customer’s wishes with Stitch Fix products to an amazing degree.
Alexnet is the leading Neural Net architecture which uses Convolution, won the Image Net LSVR competition in 2012 and is excellent for image processing.
In addition they use and other recommendation engineering algorithms and Machine Learning such as:
Collaborative filters
Mixed effects models
Naïve Bays
For a first pass so as to find the style of each customer.
Subsequently to be able to improve the Recommender after getting feedback from the customer they use algorithms such as:
Gambit neural networks
mixed effects model
So advanced Mathematical find the basic characteristics of products , with the customer’s eye , and simply mathematically calculate similarities between products .
Then they take the personification deeper by putting the human being in the game….Stylishly.
Humans process deeper variations in Recommendations, such as whether some products are too specific or too advanced, and modify the recommendations accordingly.
The CEO Katrina Lake says:
A good stylist and a good algorithm is clearly superior to a good stylist or a good algorithm.
So Business in the Digital Revolution era has shopping and customer DATA as its central feature.
Contact us now , to see how the mobiplus shopping recommendation platform can give you the above features and increase your revenue. 30%.
mobiplus member Elevate Greece
Confirmed Innovation.
Test out the Recommendation Engine for up to 20.000 euros in revenue from recommendations. No credit card required. Just enter your company information and we’ll contact you with all the details.
