How We Create Our Data Products

The data products that Environics Analytics (EA) creates, uses, and shares with clients are composed of anonymized, geodemographic and survey data at the postal code or dissemination area (DA) level in Canada and at the ZIP+6 or Block Group level in the U.S. EA obtains privacy-compliant, de-identified, anonymous, and aggregated data from reliable data partners who we rigorously vet, such as the latest census data, current economic indicators, postcensal estimates from federal and provincial governments, immigration statistics, economic data such as building permits, survey data, life stage data, specific location data and business-to-business data. Our data products include the following types of data:

Modeled Data - Data is generated from an original Census-based geographic unit (Dissemination Area or DA) and estimated, projected forward, or modelled using additional published data sources (from public and private organizations). The additional data sources are generally only published at higher levels of geography by the source but are modelled to the DA to form an EA value-added product. Examples: DemoStats, HouseholdSpend, WealthScapes, AgeByIncome.

Survey Data - Data is generated from private or syndicated survey houses that are usually only published and available from the syndicated survey house at large geographic areas (Province or Census Metropolitan Area - CMA) but are projected by EA onto the Census-based geography using a variety of techniques to form an EA valued-added product at the DA level. Examples: Vividata, Numeris, AskingCanadians.

Third-Party Sources – Third –party sources are used in the creation of some of our data products which includes privately sourced data sets that are aggregated to the DA by the data provider and undergo minor value add by EA (e.g. re-estimation of local area from higher geographic levels) before being released by EA (e.g. Equifax, Experian VIO).

Segmentation Systems - Neighbourhood Segmentation systems form a special category. These typology systems classify neighbourhoods into common segments based on various input data previously estimated at the DA or local delivery unit (LDU) level. Survey data are used as dependent sets to ensure maximum differentiation of average behaviours across segments. However, survey data are not used to determine the actual segment of a neighbourhood. Example: PRIZM^® Segmentation.

Mobile Movement Data - These data are developed using as inputs de-identified GPS ‘ping’ level data sourced from four independent aggregators. Each aggregator meets our data quality and privacy compliance standards, i.e. the app user has enabled location tracking. Large feeds are received daily. EA assigns devices to the nearest postal code to create aggregates, allowing us to infer the evening and daytime locations using the most recent three-month period. The data are used as inputs to models that normalize and weight to the general population. The models combine these summarized and de-identified movement data with other EA databases to create an aggregated estimate of anonymized data patterns for dissemination areas and postal codes. Examples: VisitorView, FootFall.