We model and interpret consumer behaviors in fine detail, right where they live
thanks to algorithms combining Geostatistics and AI
LOCAL MODELING
Exclusive surveys and hybridization algorithms
In recent years, we have surveyed more than 40,000 representative French panelists as part of our GeoInsights consumer surveys.
Algorithms combining Geostatistics and Machine Learning let us hybridize this survey data with local databases (INSEE data and, where relevant, first- and second-party data).
The result? Consumer data modeled as close as possible to the reality of territories and their inhabitants: purchase intentions, consumption habits, interests, behaviors, media preferences, and more, available locally (towns, neighborhoods, individuals) and across many sectors.
ShoWhere data is in no way personal data; it is the result of modeling applied to anonymized data sources, in strict compliance with the GDPR.
Raw information
Modeled information
GEOSTATISTICS EXPERTISE
Why Geostatistics makes all the difference
Pi(w,s) = h[ n Σ j=0 wj(s) xj,i + z(s)]
Geostatistics is not just about averages computed over territories. It is, first and foremost, a mathematical, spatio-probabilistic discipline, rich in equations.
At ShoWhere, we constantly develop and refine geostatistical algorithms. They minimize the errors of local modeling built from the heterogeneous data sources we use: panel data, INSEE data, our clients' data...
Handled through SPDE and graph-based approaches, non-stationarity is at the heart of our geostatistical models: parameters vary from one territory to the next to best reflect local realities.
These results are enriched with Machine Learning methods, helping us better capture the complexity of the data and sharpen the precision of our output.
If you'd like to learn more about our models and algorithms, don't hesitate to reach out — we'd be glad to tell you more.
AI EXPERTISE
AI to fully harness the potential of ShoWhere data
Without relevant data, AI remains fundamentally limited. Beyond the user's vigilant oversight, its performance and reliability depend on the quality of the data that feeds it.
GeoInsights data — our locally modeled consumer data, enriched by complementary territorial sources (socio-demographic, behavioral) — is the bedrock of our AI, letting us produce analyses anchored in the reality of territories and behaviors.
The results are directly actionable: deepening consumer knowledge, refining investment strategies, and generating relevant advertising message recommendations.
At ShoWhere, AI is a tool for unlocking the full potential of our data, both past and present. It lets us take analysis further and generate ever more effective recommendations, at every scale: local, multi-local or national.