Picnic is the worlds fastest growing online supermarket that makes grocery shopping simple, fun, and affordable for everyone. We will show you how we transformed an already convenient shopping experience into a delightful ultra-fast shopping blitz-stop. In this talk we provide a view behind the scenes of our deep-learning based behavioral analytics and prediction engine. We will talk you through our ups-and-downs of product, category and promotional recommendations of FMCGs and do not shy away from demoing also the failures. Now we are able to predict with >95% likelihood the top 12 articles of the next order of each of our customers. We will show you how our connected devices, ubiquitous data, and a full-fledged machine-learning logistics backend fuels all our innovations and operations.
Data Science Challenge 1: Shopping challenge. How to shop for 30 items in 3 minutes while it takes for 3 items usually 30 minutes with common apps?
Data Science Challenge 2: Distribution challenge: How to delivery in 20 minute windows while the industry still struggles with 2 hour windows?
Data Science Challenge 3: Innovation challenge: How to innovate with 2000 people as you did when you were in a garage with 20?
27. The Mobile Shopping Solution - Bulk recommendations
Set of 4, 8 or 12 articles
Buy all with a single tap
1-click shopping for half of your basket
Purchase confidence >90%
Covering repetitive & boring items
28. Input Hidden Layers Output
Monday
(wk -2)
Friday
(wk -2)
Wednesday
(wk -1)
Tomorrow
Solution: Deep Recurrent Neural Network (LSTM)
Item likelihood to buy
Cat likelihood to buy
Next 7 days
Item/cat buying interval
Order history (articles, dates)
Normalized quantities
Days between orders
x 1
x 2
x 0
x 3
x 0
x 1
x 2
x 2
x 0
x2
x 1
x 1
Item - Item
relations
Item - Day
relations
Itemset - Day
relations
y2
x1
x2
x3 y3
y1
z2
z3
z1
29. Result: Big and Deep data for optimal RFM prediction parameters
65%
70%
75%
80%
85%
90%
0 20 40 60 80 100 120 140
Precision
Number of orders
Big data
(lack of depth)
Big & Deep data Deep data
(lack of breadth)
30. The Customer Service Challenge
Blitzscaling business
Ultra-personal service
Exceptional service quality
32. Automation with Natural Language Processing
Design
Ticket enrichment
Issue identification
Product
identification
Ticket routing
Proposing solutions
Customer Ticket
many more
33. Automation with Natural Language Processing
Design
Ticket enrichment
Issue identification
Product
identification
Ticket routing
Proposing solutions
Customer Ticket
many more
34. Automation with Natural Language Processing
Design
Ticket enrichment
Issue identification
Product
identification
Ticket routing
Proposing solutions
Customer Ticket
many more
35. Automation with Natural Language Processing
Design
Ticket enrichment
Issue identification
Product
identification
Ticket routing
Proposing solutions
Customer Ticket
many more
36. Automation with Natural Language Processing
Design
Ticket enrichment
Issue identification
Product
identification
Ticket routing
Proposing solutions
Customer Ticket
many more
37. Automation with Natural Language Processing
Design
Ticket enrichment
Issue identification
Product
identification
Ticket routing
Proposing solutions
Customer Ticket
many more
38. Automation with Natural Language Processing
Design
Ticket enrichment
Issue identification
Product
identification
Ticket routing
Proposing solutions
Customer Ticket
and many more
42. 1. Dream Big, Act small
2. Mission first, Data as support
3. Data Science first, AI second
4. Launch first, Scale second
5. Great products come from small teams
Learnings from a disruptive Scale-up