Making the right decisions about where to place your next store or franchise can have a huge impact on your business, especially if you get it wrong. It’s important to make these decisions with facts that are based on accurate data and precise analytics.
Hear from hands-on industry experts and learn how to build a data-driven strategy based on location data and analytics to:
The feasibility of a new store, branch location or product offeringsThe demographics of people in a location by time of dayThe movement patterns of your customers (where they come from, where they go to, how long they spend there)How to improve customer targeting and go-to-market planning (deliver the right message, in the right place, at the right time)
Powerful Google developer tools for immediate impact! (2023-24 C)
Top 10 Tips for Retail Site Selection
1. Top 10 Tips for Retail
Site Selection
Gerry Stanley
Product Management Director
Kyle Bingham
Principal Client Manager
2. Content
Data is at the centre of site selection activities. The top 10 tips cover
understanding your data inputs through to insightful uses of data to
create high value insights.
Gerry Stanley
Product Management Director
Precisely Enrich
Stacey Grant
Marketing Manager
Precisely
Kyle Bingham
Principal Client Manager
Precisely
3. The leader in data integrity
Our software, data enrichment products and
strategic services deliver accuracy, consistency, and
context in your data, powering confident decisions.
of the Fortune 100
99
countries
100 2,500
employees
customers
12,000
Brands you trust, trust us
Data leaders partner with us
3
4. Data
Integration
Data
Observability
Data
Governance
Data
Quality
Geo
Addressing
Spatial
Analytics
Data
Enrichment
Break down
data silos
by quickly
building
modern data
pipelines that
drive
innovation
Proactively
uncover data
anomalies and
take action
before they
become costly
downstream
issues
Manage data
policy and
processes with
greater insight
into your data’s
meaning,
lineage, and
impact
Deliver data
that’s accurate,
consistent, and
fit for purpose
across
operational
and analytical
systems
Verify,
standardize,
cleanse, and
geocode
addresses to
unlock valuable
context for more
informed
decision making
Derive and
visualize spatial
relationships
hidden in your
data to reveal
critical context
for better
decisions
Enrich your
business data
with expertly
curated datasets
containing
thousands of
attributes for
faster, confident
decisions
5. 60%
19%
9%
5%
4%
3%
Cleansing & Organising Data
Collecting Datasets
Modelling/Machine Learning
Other
Refining Algorithms
Building Training Sets
5
What Data Scientists
spend most of their
time on
79% of time spent
on Data Prep Source: www.forbes.com
7. Geocoding
Turn business address
information into locations
Increased accuracy = increased
alignment with other internal
and external data
#1 The quality of location information
8. Resolution
#1 The quality of location information
1 SA3
9,100 km2
38 471 population
35,559 in 2016 – 8.2%
increase between Census
periods
2 SA2s
3 Postcodes
103 SA1s
9. Resolution
#1 The quality of location information
1 SA3
10.66 km2
56,398 population
56,066 in 2016 – 0.6% increase
between Census periods
4 SA2s 4 Postcodes 103 SA1s
11. #2 Data vintages and alignment
2016 Versus 2021
Usually Resident Population
203 (9 August 2016 – 1 SA1)
5,728 (10 August 2021 – 16 SA1s)
12. #2 Data vintages and alignment
Alignment with more
frequent products
13. #2 Data vintages and alignment
Alignment and mis-
alignment
https://www.abs.gov.au/statistics/standards/australian-statistical-geography-
standard-asgs-edition-3/jul2021-jun2026/main-structure-and-greater-capital-city-
statistical-areas/changes-previous-edition-asgs
14. #2 Data vintages and alignment
Understanding the
alignment between data
and location
15. # 3 - Classify your location by a density
measure
17. Static density
#3 Classify your location by a density measure
Population Density Building Development in a Catchment Commercial Building Density
18. Selecting the best Density
Metrics to be used is
critical
What is the target – where people live
versus where people spend time?
SA1 Population Density and SA1
Population Centroid
Or
Demographic composition of visitors to
a region at different times of day/week
#3 Classify your location by a density measure
19. # 4 - Data Harmony, leverage datasets
across the business
20.
21. Leveraging internal data
• Existing locations
• Performance
• Loyalty programs
• Transaction history
Leveraging external data
• The right data
• The right resolution
• The right currency
• Alignment to existing data
#4 Data Harmony
22. Broad Access
(singing from the same hymn sheet)
• Transparency where it makes sense
• Location assessment teams
• Decision makers
• Stock/product selection teams
Ease of Use
• Visualisation of complex data
• Dashboards with layers and filters
• Interactive maps
#4 Data Harmony
25. Collection Methods
In-store survey kiosk
QR code surveys
Web/in app surveys
Email/text surveys
Survey Content Type
Net Promoter Score focused
Customer Experience/Satisfaction focused
Product/pricing focused
#5 Generate ongoing
store survey data
77%
of customers have a more favourable view of
brands that ask for and accept feedback.
Microsoft State of Customer Service Report
74%
Of Millennials receive too many emails
70% are bothered by irrelevant ones
Retail TouchPoints
26. # 6 - Understand your store maturity
before building any types of model
28. Maturity Analysis
28
Maturity is a measurement of new-store comp growth attributed to its ‘newness/attractiveness’ and an
increase in consumer awareness of the store
29. Maturity Analysis – Example
29
72.6%
90.0%
100.0%
60.0%
65.0%
70.0%
75.0%
80.0%
85.0%
90.0%
95.0%
100.0%
Year 1 Year 2 Year 3
Maturity Ramp (% Mature)
31. Maturity Analysis
31
Year Maturity Ramp
1 72.6%
2 90.0%
3 100.0%
24.0%
11.1%
Average new store growth above and
beyond mature store comp growth
32. # 7 - Understand where mobile trace data
fits in the customer data pyramid
33. Customer Data Pyramid
33
Mobile
Trace Data
Credit Card
Data
Other (e.g. in-store Post
Code capture)
Customer
Address Level
Transaction
Data
Loyalty Card
Data
36. # 8 - Set realistic expectations on cannibilisation
37. Cannibalisation
• Highly situational; can be challenging to model
• We look for patterns by store type, by density, by
market type, etc.; review against our historical rules
• On-going research
• Two main approaches (Macro & Micro)
53. Model Overfitting
53
Overfitting is a concept in Data Science, which occurs
when a statistical model fits exactly against its training
data. When this happens, the algorithm unfortunately
cannot perform accurately against unseen data,
defeating its purpose.
An overfitted model is a mathematical model that
contains more parameters than can be justified by the
data.
Common issue with Machine Learning
54. Model Overfitting
54
Overfitting is a concept in Data Science, which occurs
when a statistical model fits exactly against its training
data. When this happens, the algorithm unfortunately
cannot perform accurately against unseen data,
defeating its purpose.
An overfitted model is a mathematical model that
contains more parameters than can be justified by the
data.
Common issue with Machine Learning
55. Model Overfitting
55
Characteristics
• Outliers exist
• Explainable
• Useful on for
new sites
Characteristics
• Few outliers; happy Clients
• Extremely difficult to explain
• Not sustainable on “new”
data; observations of 1
56. Retail Modelling - Tips
56
• Choose the right model for the right situation
57. Retail Modelling - Tips
57
• Choose the right model for the right situation
• Choose the right outcome ($’s vs. Score) for the
right situation
• Consider current store count
• Consider future format
58. Retail Modelling - Tips
58
• Choose the right model for the right situation
• Choose the right outcome ($’s vs. Score) for the
right situation
• Consider current store count
• Consider future format
• Avoid an over-reliance on AI…..for now
59. Retail Modelling - Tips
59
• Choose the right model for the right situation
• Choose the right outcome ($’s vs. Score) for the
right situation
• Consider current store count
• Consider future format
• Avoid an over-reliance on AI…..for now
• Complex networks may require more complex
models/data
60. Retail Modelling - Tips
60
• Choose the right model for the right situation
• Choose the right outcome ($’s vs. Score) for the
right situation
• Consider current store count
• Consider future format
• Avoid an over-reliance on AI…..for now
• Complex networks may require more complex
models/data
• It’s OK to have outliers, especially, if you can explain
them
61. # 10 - Use multiple methods to verify new site
forecasts