The combination of new location data streams and spatial data science techniques opens up a new array of opportunities for CPG data and marketing professionals seeking where to prioritize in terms of ramping up distribution and identifying POS (points of sale).
4. CARTO — Unlock the power of spatial analysis
Replace this image
How Spatial Data
can be used to
reveal features &
areas for
successful
distribution
rollout?
5. CARTO — Unlock the power of spatial analysis
“Organic is the fastest growing sector of
the U.S. food industry”
Source: https://ota.com/hotspots
6. CARTO — Unlock the power of spatial analysis
➢ How can we identify
the hotspots?
➢ What drives the growth
in demand and the
location preference?
➢ How can we use
location
information-data to
extrapolate?
Growth in demand is affected by:
● Culture
● Socio-economic factors
● Health factors
● ?
Hotspots are related with:
● Where people spend their
money
● What are their interests
depending on the location
● What do they search online
● ?
7. CARTO — Unlock the power of spatial analysis
● How can we identify the hotspots?
● What drives the growth in demand and the location
preference?
● How can we use location information-data to extrapolate?
● Culture
● Socio-economic factors
● Health factors
● ?
● Where people spend their
money
● What are their interests
depending on the location
● What do they search online
● ?
Growth in demand is affected by: Hotspots are related with:
8. CARTO — Unlock the power of spatial analysis
POLL 1
Do you have at least one social media account?
(LinkedIn, Twitter, Instagram, TikTok, Foursquare, Snapchat, etc)
Yes
No
9. CARTO — Unlock the power of spatial analysis
79%
US Population uses social media
Source: Statistica: https://www.statista.com/statistics/273476/percentage-of-us-population-with-a-social-network-profile/
10. CARTO — Unlock the power of spatial analysis
72
Geosocial
Segments
11. CARTO — Unlock the power of spatial analysis
Spatial.ai
Geosocial Segments: behavioral segments based on the
analysing social media feeds with location information
Mastercard
Geographic Insights: providing sales-based dynamics of a
location with indices measuring the evolution of credit card
spend, number of transactions, average tickets, etc.
happening in a retail area over time
Dstillery
Behavioral Audiences: audiences derived from online
behaviors
Pitney Bowes
Points of Interest: database with the location of businesses
and other points of interest categorized by classes and
industry groups
AGS
Sociodemographics: basic socio-demographic and
socio-economic attributes estimated at current year and
projected 5 years into the future
What Data have
we used?
In the new millenia people tend to
express their interest and
preferences in social media.
People use the internet search
engines to find whatever they want.
Can we use information from social
media and internet to identify the
hotspots apart from socio economic
factors?
12. CARTO — Unlock the power of spatial analysis
Data Sources
Behavioral
Geosocial Segments: behavioral segments based
on the analysing social media feeds with location
information
Behavioral Audiences: audiences derived from
online behaviors
POI’s
POIs: database with the location of
businesses and other points of interest
categorized by classes and industry groups.
Demographics
Sociodemographics: basic
socio-demographic and socio-economic
attributes estimated at current year and
projected 5 years into the future
Geographic Insights: providing sales-based dynamics
of a location with indices measuring the evolution of
credit card spend, number of transactions, etc.
happening in a retail area over time
Financial
COMMERCE
PEOPLE
Physical Digital
13. CARTO — Unlock the power of spatial analysis
🐶 🐕 🐾
#Puppylove
#Dogsofinstagram #fur
Woof #mansbestfriend
#Dogmom
#Furbabies
Walks #dogtoy
Clustered
Text Data
#dogbreeds Grooming
#Puppylove +100s more
Kong
Geographic
Segment
19. CARTO — Unlock the power of spatial analysis
1. Average ticket size in Grocery Stores based
on Mastercard data
2. Organic food has potentially a higher
demand via the exploration of social media
posts (using Spatial.ai geosocial
segmentation) and internet search
behaviours (with Dstillery's audience data)
How can we identify the
hotspots?
Built a classifier which considers the socio
economic and geosocial segments (Spatial.ai
data) to identify which features are
“responsible” for the selection of the
“targeted” areas
{Selected areas} = {Mastercard ∩ {Spatial.ai ∪ Dstillery}}
What drives the growth in
demand and the location
preference?
20. CARTO — Unlock the power of spatial analysis
Resulting Areas New York
Link
21. CARTO — Unlock the power of spatial analysis
Resulting Areas Philadelphia
Link
22. CARTO — Unlock the power of spatial analysis
Exploring features and Characterizing the selected areas
● Perform t-test to identify which features are “different” between selected and the rest of the areas
● Further reduce the dimension of Geosocial segments, see the differences between the selected and
non-selected areas
23. CARTO — Unlock the power of spatial analysis
Building a
classifier
For the remaining features:
● Upsampling the imbalance
dataset.
● Random forest Classifier.
● Output the significance of
each feature to whether or
not a block should be
labelled as “targeted”.
Identification of the driving factors
24. CARTO — Unlock the power of spatial analysis
Main driving factors for New York
25. CARTO — Unlock the power of spatial analysis
Main driving factors for
Philadelphia
26. CARTO — Unlock the power of spatial analysis
It’s time for a real world example!
27. CARTO — Unlock the power of spatial analysis
Identifying twin
areas in different
cities
Example
Available data:
● Per capita income (projected, five years)
● Average household Income (projected, five years)
● EB03_lgbtq_culture
● ED09_hops_and_brews
● ED08_wine_lovers
● ED04_whiskey_business
● Median household income (projected, five years)
● ED02_coffee_connoisseur
● LEGAL SERVICES
● ED01_sweet_treats
Selected block in New York
28. Thanks for listening!
Any questions?
Request a demo at CARTO.COM
Lyden Foust
CEO of Spatial.ai // lyden@spatial.ai
Argyrios Kyrgiazos
Data Scientist at CARTO // argyrios@carto.com
29. CARTO — Unlock the power of spatial analysis
{Selected areas} = {Mastercard ∩ {Spatial.ai ∪ Dstillery}}
Methodology
Our analysis follows two main steps:
● Identification of target areas with high potential for a successful rollout of organic products.
■ Identification of areas with higher average ticket size in Grocery Stores based on Mastercard data
■ Identification of areas where organic food has a potentially higher demand via the exploration of
social media posts (using Spatial.ai geosocial segmentation) and internet search behaviours (with
Dstillery's audience data)
■ Intersection of the areas identified in the above two steps; these will be the resulting selected target
areas for the reminder of the case study.
○ Analysis of the different factors that characterize and have driven the selection of the target areas, build a
classifier
● Identification of twin areas in San Francisco based on those selected in New York and Philadelphia
30. CARTO — Unlock the power of spatial analysis
Study of people based on
where they live.*
Study of people based on what
they do.
The Traditional Data Landscape
Government mandated
survey.
Census Data Psychographic Data True Human Behavioral Data
*Harris, Sleight, Webber. Geodemographics, GIS and neighborhood targeting. Wiley, 2005