This document provides an overview and agenda for a presentation on using graph databases like Neo4j for retail applications. The presentation covers introducing graph databases and Neo4j, discussing retail data types, and demonstrating use cases for customer 360 views, recommendations, supply chain management, and other areas. Case studies are presented on using Neo4j for real-time recommendations at a large retailer and real-time promotions at a top US retailer. The document concludes with an invitation for questions.
Graphs in Retail: Know Your Customers and Make Your Recommendations Engine Learn
1. Graphs in Retail
Know Your Customers and Make Your Recommendation Engine Learn
Joe Depeau
Sr. Presales Consultant, UK
14th August, 2019
@joedepeau
http://linkedin.com/in/joedepeau
2. • Introduction to Graphs and Neo4j
• Retail Data Overview
• Use Cases
• Customer 360 View
• Recommendations (with demo)
• Logistics
• Supply Chain Management
• Others …
• Q&A
2
Agenda
7. 7
Car
DRIVES
name: “Dan”
born: May 29, 1970
twitter: “@dan”
name: “Ann”
born: Dec 5, 1975
since:
Jan 10, 2011
brand: “Volvo”
model: “V70”
Anatomy of a Property Graph Database
Nodes
• Represent the objects in the
graph
• Can be labeled
Relationships
• Relate nodes by type and
direction
Properties
• Name-value pairs that can go on
nodes and relationships.
LOVES
LOVES
LIVES WITH
OW
NS
Person Person
9. 9
Some Examples of Typical Retail Data
Event DataProduct
Data
Customer DataOrganisational
Data
3rd Party Data
Documentation
Facilities
Processes
Systems and
Databases
KPIs and Reports
Personal Data
Customer
Relationships
Documentation
Processes
Brand Data
Product
Hierarchy
Pricing Data
Clickstream Data
Searches
Customer Contact
Social Media
Market Data
Organisational
Hierarchy
Purchase History
Supply Chain Data
Supplier Data
Logistics Data
Inventory Data
Local Data
11. Customer 360 Example Graph
Organisational Data
Customer Data
Product Data
Event Data
3rd Party Data
Supply Chain Data
11
12. 12
Customer 360 Graph Uses
• Can I use the graph to help me improve customer
experience?
• Can the graph help me determine Customer Lifetime
Value (CLV)?
• Can I spot and prevent churn using a graph?
• Can the graph help me spot influencers in my
customer base?
Yes!
Yes!
Yes!
Yes!
14. 14
Product
Recommendations
Effective product recommendation
algorithms has become the new
standard in online retail — directly
affecting revenue streams and the
shopping experience.
Logistics/Delivery
Routing recommendations allows
companies to save money on
routing and delivery, and provide
better and faster service.
Promotion
recommendations
Building powerful personalized
promotion engines is another area
within retail that requires input from
multiple data sources, and real-time,
session based queries, which is an
ideal task to solve with Neo4j.
Today Recommendation Engines are at the
Core of Digitization in Retail
16. 16
Recommendations Graph Uses
• Can I use the graph to help me improve sales with
better recommendations?
• Can the graph help me make real-time
recommendations across channels?
• Can I integrate my recommendations graph with AI
and Machine Learning techniques?
• Can the graph help me with other types of
recommendations besides products?
Yes!
Yes!
Yes!
Yes!
18. Supply Chain Example Graph
Organisational Data
Customer Data
Product Data
Event Data
3rd Party Data
Supply Chain Data
18
19. 19
Supply Chain Graph Uses
• Can I use the graph to help me improve my ordering
and procurement processes?
• Can the graph help me save money on orders?
• Can I optimize my inventory using a graph?
• Can the graph help me with comparative analysis of
my suppliers and their products?
Yes!
Yes!
Yes!
Yes!
20. 20
Other Retail Graph Use Cases
• Identity and Access Management
• Infrastructure and Network Management
• Master/Meta-data Management
• Regulatory Compliance (i.e. GDPR)
23. 23
Case studySolving real-time recommendations for
the World’s largest retailer.
Challenge
•In its drive to provide the best web experience for its
customers, Walmart wanted to optimize its online
recommendations.
•Walmart recognized the challenge it faced in
delivering recommendations with traditional relational
database technology.
•Walmart uses Neo4j to quickly query customers’ past
purchases, as well as instantly capture any new
interests shown in the customers’ current online visit
– essential for making real-time recommendations.
Use of Neo4j
“As the current market leader in
graph databases, and with
enterprise features for scalability
and availability, Neo4j is the right
choice to meet our demands”.
- Marcos Vada, Walmart
• With Neo4j, Walmart could substitute a heavy batch
process with a simple and real-time graph database.
Result/Outcome
24. 24
Top Tier US Retailer
Case studySolving Real-time promotions for a top
US retailer
Challenge
•Suffered significant revenues loss, due to legacy
infrastructure.
•Particularly challenging when handling transaction
volumes on peak shopping occasions such as
Thanksgiving and Cyber Monday.
•Neo4j is used to revolutionize and reinvent its real-
time promotions engine.
•On an average Neo4j processes 90% of this retailer’s
35M+ daily transactions, each 3-22 hops, in 4ms or
less.
Use of Neo4j
• Reached an all time high in online revenues, due to
the Neo4j-based friction free solution.
• Neo4j also enabled the company to be one of the first
retailers to provide the same promotions across both
online and traditional retail channels.
“On an average Neo4j processes
90% of this retailer’s 35M+ daily
transactions, each 3-22 hops, in
4ms or less.”
– Top Tier US Retailer
Result/Outcome
25. 25
Case Study : Wobi
uses Neo4j to enable
‘Whole Customer
Understanding’
The World’s Leading Graph Database
CASE STUDY
www.neo4j.com
Wobi
Price Comparison Site Wobi Builds
‘Whole Customer Understanding’
with Neo4j
The success of price comparison websites rests on their ability to make
customers compelling ‘value offers’ – and to do that they need to capture,
organise and instantly analyse masses of customer data. Israel-based Wobi
has achieved that aim of ‘whole customer understanding’ using Neo4j.
The Company
Founded five years ago, price comparison website Wobi is already one of Israel’s best-known com-
panies. Owned by the White Mountain investment group, Wobi has over 500,000 customers and
millions of site visitors every month, who use Wobi to compare and choose their pensions and car,
home, mortgage and travel insurance. Wobi has around 100 staff and, bolstered by a high-profile
TV advertising campaign, will expand further this year by launching a banking and finance compar-
ison service.
The Challenge
Wobi aims to give its customers best ‘value offers’, and to do that it needed to build a detailed
picture of each customer and their full financial situation – savings, pensions, insurance policies,
accounts and family background.
As Chief Technology Officer, Shai Bentin, explained: “We look to give our customers great value
offers and so, as our CEO says, we want to look at the customer’s account in such depth that we
can tell them they have a leak in their house because they have been paying more for their water
every month! That’s the idea...we can offer to move a customer from, say, one phone company to
another that better suits their needs – and we can read that information off their account, their
credit account and the way the customer behaves.”
To achieve that level of understanding, Wobi needed a single customer database where it could
rapidly drill-down into each individual’s history and add new information on the fly.
It faced two key issues. “One is that we need to extract a lot of customer information very, very fast
from the database,” Shai said, “and the second is the way we get the information. It’s a tree-like
structure – under each customer will hang a lot of information, and for performance we needed to
pull up all that information at once.”
When Wobi began searching for the ideal database, it realised “that structure really suited working
with Neo4j”, Shai said, because Neo4j organises data into ‘nodes’ and ‘relationships’. This enabled
Wobi to define its customers as ‘nodes’, and to hang off them every piece of information relating to
that customer as ‘relationships’.
Shai explained: “Instead of having to break up our data into tables like with an SQL database and
make thousands of joins, with Neo4j we could just ‘save the tree’ and do a single look-up to the
person, to grab everything at once.”
The Strategy
Neo4j is now Wobi’s core customer database, sitting at the heart of a network of around 20 servers,
with a team of five people doing Neo4j development and testing work.
Wobi started using Neo4j after coming across the product by chance. Shai explained: “I felt that
working with a normal SQL database would be too much work for us, and I actually started out look-
ing for object databases, because our programming language is Java which is very object oriented.
Then I stumbled on Neo4j – we tried it out and it worked for us.”
INDUSTRY
Finance
USE CASE
Graph-Based Search/
Recommendations
GOAL
Make customers ‘value
offers’ based on in-depth
understanding of their current
financial situation and needs
CHALLENGE
Rapidly analyse large volumes
of ‘whole customer’ information
SOLUTION
Store all customer data into
Neo4j database
RESULTS
– Data on half-a-million
customers is accessed
exponentially faster
– All data consolidated in
Neo4j for ‘whole customer
understanding’