Abstract: To determine an impression metric for an organization, a server device generates a statistical model for estimating the impression metric using machine learning techniques. The server device obtains training data for the statistical model by randomly selecting geographic locations within a geographic area and determining the number of users eligible to receive a particular type of advertisement for each randomly selected geographic location. For example, a user may be deemed eligible when displaying the geographic location via a mapping application. When an organization requests an estimate of a number of impressions for an advertising campaign, the server device applies data included in the request (e.g., the time period for the advertising campaign, the number of organization locations, identifiers for the organization locations such as geographic coordinates or an address, etc.) to the statistical model to estimate an impression metric for the organization.

Abstract: To determine an impression metric for an organization, a server device generates a statistical model for estimating the impression metric using machine learning techniques. The server device obtains training data for the statistical model by randomly selecting geographic locations within a geographic area and determining the number of users eligible to receive a particular type of advertisement for each randomly selected geographic location. For example, a user may be deemed eligible when displaying the geographic location via a mapping application. When an organization requests an estimate of a number of impressions for an advertising campaign, the server device applies data included in the request (e.g., the time period for the advertising campaign, the number of organization locations, identifiers for the organization locations such as geographic coordinates or an address, etc.) to the statistical model to estimate an impression metric for the organization.