Slide 1
Slide 1 text
No content
Slide 2
Slide 2 text
39 United
States
Patents
Erdös
Number 3
Expt. &
Theoretical Physics
Applied &
Computational
Mathematics
Began Career at Microsoft Research
9 Startups incl. INRIX, Amplero
Consultant to >40 global brands
Slide 3
Slide 3 text
No content
Slide 4
Slide 4 text
Data-driven approaches work, but the
impact can be unexpected!
The impact is directly related to the
metrics you choose to optimize.
For long term positive impact the
customer experience and your business
metrics must align rather than compete.
Slide 5
Slide 5 text
No content
Slide 6
Slide 6 text
No content
Slide 7
Slide 7 text
No content
Slide 8
Slide 8 text
8
A NEW STORE EVERY DAY
INSPIRED, DISCOVERY-DRIVEN EXPERIENCE
HIGHLY CURATED
A DAILY DESTINATION
MASSIVELY PERSONALIZED APPROACH
Slide 9
Slide 9 text
V
Slide 10
Slide 10 text
•
•
•
•
Slide 11
Slide 11 text
Machine Learning
Sense Think Act
Feedback/Metric
Artificial Intelligence
Slide 12
Slide 12 text
Optimizing for Likelihood to
Buy
Optimizing for Likelihood to
Spend
Likelihood of Purchase
Order Value
Profitability
Slide 13
Slide 13 text
•
•
*Example Data Shown
Slide 14
Slide 14 text
•
•
Percent of Sort Rankings for which event is in Top 6 Distribution of how event is sorted across all
customers
Example Data Shown
Slide 15
Slide 15 text
•
Cumulative Click-Throughs
Day 1 Day 2 Day 3
Example Data Shown
Slide 16
Slide 16 text
Be deliberate in considering the
business metric you are trying to
impact
If you can’t directly connect your
experience outcomes to the business
metric find a proxy
Continually experiment at low
business impact before you dive all
the way in!
Slide 17
Slide 17 text
No content
Slide 18
Slide 18 text
USE CASES
Slide 19
Slide 19 text
TYPICAL SOCCER MATCH IN NUMBERS
•
•
•
•
Slide 20
Slide 20 text
SCORING GOALS
•
•
•
•
•
•
Slide 21
Slide 21 text
SHOTS METRICS
•
•
•
Slide 22
Slide 22 text
PRIME AREA FOR GOAL SCORING
Slide 23
Slide 23 text
CHANCES
•
•
•
Slide 24
Slide 24 text
DIRECT PLAY
•
•
Slide 25
Slide 25 text
POSSESSION
Text
Text
• Possession – Advance the ball into
opponent half, opponent 3rd and try to find
a team mate in the high-value shooting
zone.
• Higher quality chances, but needs skill and
better play to advance the ball through
possession
Slide 26
Slide 26 text
PLAYING STYLES AND PLAYERS
•
•
•
•
Slide 27
Slide 27 text
SELF-SCOUTING – KNOWING WHO WE ARE
•
•
•
•
Slide 28
Slide 28 text
EXAMPLE: ATTACK PAZ (PRIME ASSIST ZONES)
Slide 29
Slide 29 text
ALL THIS SEEMS INTUITIVE AND NOT ROCKET SCIENCE
• ☺
•
Slide 30
Slide 30 text
INTEGRATION INTO TRAINING DRILLS
•
Slide 31
Slide 31 text
METRICS
•
•
•
•
•
•
•
•
Slide 32
Slide 32 text
SCOUTING PLAYERS
•
•
•
•
•
•
•
Slide 33
Slide 33 text
IN SUMMARY
•
•
•
•
Slide 34
Slide 34 text
No content