Luminate tracks music consumption activity.
We do this by ingesting reported activity from thousands of data sources.
We use that data to construct and maintain a musical entity database.
This allows us to attach metadata and activity to artists, songs and albums.
Our model is upheld industry-wide as a standard metric of music performance.
Read on to learn more about our methodology and how we apply it within CONNECT. For specific questions, refer to Methodology FAQs.
CONNECT aggregates data related to music consumption: all the different ways people listen to music. We track consumption activity within four main categories, called Activity Types: Streams, Product Sales, Song Sales and Airplay.
These activities combined represent the vast majority of worldwide music listenership.
Streams
All music streaming
Product Sales
Physical and digital sales of albums, EPs and singles
Song Sales
Digital song sales
Airplay
Satellite and terrestrial radio play measured by spins and audience
Some Activity Types can be broken down into additional categories. We segment Streams by commercial model (ad-supported or premium), Product Sales by purchase method (online or storefront) and so on. See below or refer to our glossary of terms for more.
Read more: Activity Types: Tracking music consumption methods
Streams, sales and spins are unequal measures of listener engagement and economic value. So we use a weighted activity metric to compare disparate units and present a more accurate big-picture view. Our formulas convert Streams, Product Sales and Song Sales into equivalent units that can be tracked, ranked, and measured.
We (mainly) use Album Equivalent and Stream Equivalent units to measure music performance and rank titles on the charts.
Weighted streaming, digital and physical sales figures converted to 1 album sale unit.
Weighted streaming and digital song sales figures converted to 1 premium stream unit.
Luminate’s music consumption equivalents, developed and refined over more than 30 years, are the industry standard for evaluating music popularity and performance. Our framework for measuring music consumption sets industry benchmarks, helps determine market share and powers the Billboard charts.
Read more: Equivalent weighting: Measuring music consumption
CONNECT compiles over 23 trillion annual data points related to thousands of artists and millions of songs. Our data comes to us directly from more than 500 partners across entertainment, retail and technology.
On the back end, the CONNECT platform automatically ingests, validates and processes the data. Then, CONNECT applies our programming logic to build a database. We can then attach activity to entities and apply equivalent formulas to measure consumption.
In addition to consumption data, we capture metadata—release date, country of origin, language, format and other identifying information—for artists, songs and release groups (albums). Our metadata allows us to group related items and provide advanced analysis based on attributes like genre, location, artist type, release age and more.
We track activity data and metadata for millions of musical entities. A musical entity can be an artist, song or album. Within those broad descriptors, CONNECT recognizes five musical entity types: Recordings, Songs, Products, Releases, Release Groups and Artists.
Albums and Songs are both umbrella categories containing lower-level subcategories.
Artists
No subcategories
Albums
Release Groups
Releases
Products
Songs
Songs
Recordings
Albums (Release Groups)
“Album” is an imprecise term. We use Release Group and Release to cover every multi-track musical product—mainly albums, EPs and singles.
A Release Group comprises every version and format of an album. These include Releases and Products.
Release Group
↳Release
↳Product
The Release level differentiates between versions of an album (i.e. initial release vs. expanded, remastered or remixed). If an album has been reissued with the same title and tracklist, but now includes noticeable differences such as remastering or bonus songs or a second disc of unreleased demos, each of those versions is one Release.
The Product level refers to individual barcodes (ICPNs). Most Releases are sold in more than one format: CD, vinyl, digital album, retailer exclusive, alternative cover art, limited-edition colored vinyl, et cetera. Each of these is its own Product.
Songs
Songs (sometimes called tracks) are usually identified by title and artist. But one Song might have multiple Recordings.
A Recording is just that: one recorded version of a song. It can be a studio version, a remaster, a re-recording, a live version, a demo or even a remix—as long as it’s the same song. Every Recording is identified by a unique code called an ISRC.
Song
↳Recording
↳Recording
↳Recording
We construct our database “from the ground up.” Providers share metadata and performance data for Recordings and Products. We then apply our grouping rules and use programming logic to sort entities into higher-level categories.
Musical entities (see above) show up in CONNECT on Dashboards. Think of a Dashboard as a visual representation of an entry in the CONNECT database.
Activity tracking starts at the Recording (ISRC) level for streams and song sales. Our partners share streaming and sales data for individual ISRCs. We attach metadata to ISRCs, which allows CONNECT to group Recordings into Songs and assign them to Artists.
We track album, single and EP sales at the Product (ICPN) level. Retailers report product sales by barcode/ICPN; each ICPN submission includes metadata that helps us compile Products into Releases and Releases into Release Groups.
Read more: ISRC, ICPN, ISNI: External IDs and product codes
Tracking artist metadata allows us to disambiguate artists, display key information on Artist Dashboards, attribute Recordings and Products and calculate overall artist metrics. We receive artist data from our data partners or as manual submissions.
The above process allows us to maintain our database and populate the platform at scale. But some data shares contain errors. If we receive conflicting metadata from multiple sources, we apply programming logic to select the best option for each field. When necessary, our Operations team manually reviews and corrects data.
We accept submissions, updates and corrections to help build our database and improve accuracy. And we allow individual registration (with verification) of new artists/titles or those not covered by our data sources.
Read more: Music registration: Add entries to our database
Our methodology rests on a number of specialized concepts. We also use some terminology of our own.
Review (or bookmark!) these Knowledge Base links for quick reference. Log in with your CONNECT subscription email to access most Knowledge Base articles.
Glossary: Acronyms, key terms and definitions
ISRC, ICPN, ISNI: External IDs and product codes
Data Controls: Activity Types, Breakout Views and filters
About Genres: How Luminate classifies music
Methodology FAQs: Our formulas, data sources, and eligibility criteria
Equivalent weighting: Measuring music consumption
How to CONNECT: All tutorials, walkthroughs and video demos