Data References
Rigorous data. Transparent methodology.
Every GF Score is built on verified, multi-source data, and we publish the methodology because you should be able to check our work — every input traced to its source, every weight yours to inspect.
REF-PILLARS-01
The four data pillars
Foot Traffic
Unacast
Demographics
U.S. Census + Esri enrichment
Traffic VPD
StreetLight Data + DOT counts
Competition & POI
Dataplor + GrowthFactor proprietary
Data Pipeline
From search to scored location in seconds
Four stages turn every search into a rigorous, defensible score — hitting our data providers in real time, with full traceability at every step.
- 4
- Data sources
- ~10 sec
- Address to scored report
- 370M+
- POIs accessible
- Weekly
- Fastest data refresh
Query
Every search triggers real-time API calls to our four data providers for the requested trade area — no stale snapshots or pre-built caches.
On demandNormalize
Provider responses are standardized to H3 hexagonal grids, deduplicated, and validated against census boundaries at query time.
Real-timeScore
The lens-based scoring model — five lenses by default, configurable per brand — weights each dimension and produces a composite GF Score with per-lens sub-scores.
Real-timeDeliver
Scored locations surface in the platform dashboard, shareable maps, and analyst reports in seconds.
SecondsData Sources
Four data pillars behind every score
Each data source is independently validated, continuously refreshed, and calibrated against ground-truth observations.
Foot Traffic
Device-level visit data drawn from Unacast's panel of 1B+ monthly devices. Machine-learning models validated at 91.6% R² against ground truth, with duplicate and fraudulent signals cut by up to 65%.
Foot Traffic
Panel scale & validation
1B+
Monthly devices
91.6%
Validation R²
65%
Duplicates cut
Source: Unacast methodology documentation
Demographics
US coverage & depth
2,000+
US variables
13M+
Business records
67
Tapestry segments
Source: Esri Business Analyst, U.S. Census ACS
Demographics
Population density, household income, age distribution, and consumer spending — down to the Census block-group level. Esri layers 2,000+ US demographic variables on top of ACS data with current-year projections refreshed quarterly.
Traffic VPD
Vehicles per day derived from ~40B anonymized mobility probes per month. Calibrated against 6,600+ US permanent counter stations at R² 0.98 vs. ground-truth AADT, with 100% coverage of US Traffic Analysis Zones.
Traffic VPD
Probe volume & accuracy
40B
Monthly probes
0.98
Counter R²
100%
US TAZ coverage
Source: StreetLight Methodology White Paper
Competition & POI
POI coverage & freshness
370M+
POIs indexed
15,000+
Brands tracked
Weekly
Verified refresh
Source: Dataplor coverage data
Competition & POI
Competitor locations and points of interest from Dataplor's 370M+ POI database, refreshed weekly with in-market human verification. Our proprietary layer adds closure signals and lease-up tracking.
Methodology
Five-Lens Scoring Model
Every location receives a composite GF Score from 0 to 100, computed as a weighted average of independent lenses — five by default, with lenses and weights configurable per brand. Each lens produces its own sub-score, so you can see exactly what's driving the composite.
Visibility
Storefront exposure, signage opportunities, and impressions from vehicle and pedestrian traffic.
Demographics Fit
Household income, population density, age distribution, and alignment with target customer profile.
Market Potential
Trade zone population, retail spending potential, and market growth trajectory.
Traffic & Accessibility
Daily vehicle traffic, parking availability, and ingress/egress accessibility for car-based shoppers.
Competition
Competitor density, complementary anchors, and co-tenancy patterns that drive foot traffic.
Every score in the platform shows its lens weights, so nothing is hidden behind a composite. Weights are configurable per brand, and Enterprise onboarding sets them with your team.
REF-MODEL-01
Sample Output
Lenses Breakdown
Analyst Team
A data scientist who knows your markets
GrowthFactor Labs builds your custom forecasting model alongside a named data scientist, and Enterprise onboarding configures your scoring criteria with your team, so you understand every weight and can explain every number yourself.
REF-TEAM-01
What that looks like
Scoring model, configured together
Lenses and weights set with your team at Enterprise onboarding, not defaults handed down.
Custom forecasting models
Built alongside your named data scientist, calibrated to your fleet.
Every weight inspectable
Each score in the platform shows its lens weights — nothing hidden behind a composite.
See it on your own sites
See the data on a site you're debating
Book a demo and bring a recent site. We'll score it live and walk through the inputs lens by lens, on your trade area.