The presented article is devoted to the theoretical and methodological description of the DATA FUSION panel used in media measurements of rating indicators of television viewing by the TV audience. This tool for fixing rating indicators (rating, coverage, share, matching index (affinity), average viewing time, etc.) is used in modern sociological media research to obtain representative audience data on television consumption. In terms of structure and content, the DATA FUSION panel is an integrated array of empirical "big" data (BIG DATA) of STB set-top boxes (set-top boxes) and empirical indicators of a TV panel (panel volume - 600 households representing the urban population of the Republic of Belarus), equipped with people metric (PM) devices for fixing television viewing. This panel allows, on the one hand, to clarify and supplement the audience data obtained by means of pilmeters (passive audience measurement device), on the other hand, to endow with socio-demographic characteristics "big data" (BIG DATA) provided by set-top boxes (approximately 2,000,000 ZALA set-top boxes , "Beltelecom"). In these contexts, the purpose of this article is a conceptual analysis of the mechanism for implementing the methodological tools of the DATA FUSION panel in the media dimensions of television. The object of research interest is the DATA FUSION panel as a tool for fixing TV viewing ratings. The presented article describes the concept of methodological tools of the DATA FUSION panel, which is used in test mode by the research company MediaMeter (Minsk, Republic of Belarus) when fixing the rating indicators of the Belarusian television audience.