8. Our Agenda
• The Landscape of Big Data
• 5 Key Principles to Really Benefit from Working
With Big Data
• How We Apply These in Our Work
• Some More Opportunities for Transformation
@rolfeswinton
20. And the Next Step Change…
• The fist quantum computer
now on-line = 50,000+ servers
• Wearable computing = near
perfect information on
consumer behavior
@rolfeswinton
22. Big Data Laws #1
Start with the pain in mind
What is the specific question you need to
answer?
23. Big Data Laws #2
• Data needs to come together in one place
– Single CRM / customer identifier
– Personally Identifiable Information (PII)
– Unified data structure across business (silos)
– Sensor data
– Social data, photos, messages, etc.
– Small Data + Big Data
• And plan for explosive growth in data volumes
as you unify it
24. Big Data Laws #3
• Bring specific questions but be ready for
surprising answers, and the need to change the
question
– Creativity & science
– Machine & human
– Variety of ways to explore the data (visualisation)
“The racing technology on the yachts competing for the
2013 America’s Cup will be the most advanced ever”
- The Wall Street Journal MarketWatch
25. Big Data Laws #4
• The greater the speed of analysis, the greater the
predictive value
• But usually means rethinking current business processes…
26. Big Data Laws #5
• Understand Why You Have the Data You Have
– You have the capacity to visualise people’s lives
– Better be able to justify it to your customers and
to regulators
– Better be able to understand what’s worth
keeping and what you need to get rid of
30. #1 – What’s the Pain
How to Increase Profitable Sales?
@rolfeswinton
31. #1 – What’s the Pain?
The Hypothesis
Attract New Most
Customers Difficult
Easiest,
Sell to Existing Fastest,
Customers Least Risky
Sell Existing Types Introduce New Types
of Merchandise of Merchandise
@rolfeswinton
32. #2 Bring Data Together
Big Retail Data Sets to Fuse
Customer
consumer feeback
$
Channel Concept Identification of the
Shopper, Inven
tory Specific
& Opportunities for
Financial Growth
Data
Categories Competition
@rolfeswinton
33. #2 Bring Data Together
Data From the Entire Path to Purchase
Ad GPS/Triangulation Store-level Fixture Level
Analytics Location and Mobile Intelligence Intelligence
Behavioral Analytics WiFiData Mobile Data
Inventory Data POS Data
@rolfeswinton Financial Data
34. #3 Be Ready for Surprises
The Scale of Opportunity
The total potential
customer spend that
can be addressed by the
retailer
@rolfeswinton
35. #3 Be Ready for Surprises
When is the Optimal Time to Reach Digital Shoppers?
@rolfeswinton
36. #3 Be Ready for Surprises
Why Does WiFi in Stores Drive Increase Sales?
@rolfeswinton
37. #3 Be Ready for Surprises
Relative Opportunity by Customer Segment
Proprietary
analytics to
identify and
quantify specific
customer
segments for
targeting that have
the greatest
potential
@rolfeswinton
38. #3 Be Ready for Surprises
Understanding Right Pricing
PRICE POINT NOT
A.S.P. REPRESENTED
12%
PERCEIVED PRICE POINTS PERCEIVED Identify Where
10% CHEAP TOO EXPENSIVE AT CLIENT
Additional Options
8% Are Justified Or
6% Where The Category
4% Needs To Be Edited
2%
– Can Be Done By
Channel
0%
$0 $100 $200 $300 $400 $500 $600 $700 $800 $900
}
Inventory
Concentration
@rolfeswinton
39. #3 Be Ready for Surprises
Top 5 Reasons Why Customers Buy at Major Competitor
Brands I Want $11.1M
More Choice
Sales/Promotions Not Having the Right Brands at
Our Client Costs the Company
Good Return Policy $11.1M in Lost Sales to its Major
Competitor
Usually In-Stock
$0 $2.5 $5.0 $7.5 $10.0 $12.5
Lost Sales Opportunity ($Millions)
@rolfeswinton
40. #3 Be Ready for Surprises
What’s the Optimal Offer to Deliver to Specific Shoppers at
Specific Times?
How to
2 for 1 10% off Use the
Product
@rolfeswinton
41. #3 Be Ready for Surprises
What is the Opportunity Inside the Store?
Lack of Clear Information
Hierarchy & Poor
Customer Circulation Costing
$78M = Strategic information
delivery
―Showrooming‖ via
competitor sites
costing$132M = mix of Long Checkout Lines
smarter bundled offers Costing $275M = automated
& in-store help / support staff triggers to add tills or
mobile checkout staff
$320M opportunity to capture
sales from one key competitor
via targeted offers optimized
through real-time A/B testing
@rolfeswinton
42. #4 Speed = Predictive Value
Applying Real-time Analytics
Strategic problem identified:
e.g. Long checkout lines cost
the company $275M annually
Dashboards set-up to report
5
daily checkout line avg. wait
4
times to operations management
3 Top Quartile
2
Bottom
1 Quartile Messages sent to store
0 managers in real-time
Jan Mar May Jul Sep Nov when long queues are
Feedback anticipated
loop
@rolfeswinton
43. Real Impact
This Retailer increased sales by
over 20% – an improvement of
hundreds of millions of dollars –
with an increase in gross margin
Only part of this implemented to
date…
@rolfeswinton
44. #4 Speed = Predictive Value
Create More Shopper Value
• Manage the
appropriate delivery of
– Digital product information
– Coordinated item suggestions
– Targeted promotional coupons
– YouTube and/or other video content
– Customer reviews of items
– Real-time customer feedback
– Help available online
– Stock checking
– Online ordering
– Auto negotiation tools
– Instant ability to request live help
– And more…
@rolfeswinton
46. What’s it Worth?
―The Most Profitable Customer is
the Omni-channel Customer‖ — Forrester
Relative difference in sales by customer
$4 : $1
Omni-channel
customer
Single channel
customer
@rolfeswinton
CaloriesMovementSleep patterns – measurement & coachingHeart RateWeightFatAir QualityHow you spend your digital life? Other sensors to be embedded in anything (tooth brushes, water bottles, etc.)
400 Million
Northrop Grumman has now gone live with the first quantum computer.
Same satelites / different sensor / better processing – compute power at the edge
Insurance
Aetna has 5 key metrics for determining changes in patient health
Sensors + Big Data + Fast analytics meant the bullet trains stopped 80 seconds before the earthquake hit saving thousands of lives