Labels and artists have to trust stores when it comes to collecting their revenues. Error and attempted fraud are hidden inside huge amounts of data. We want to change this.<\/p>\n
Settlement of music streaming and download revenues is becoming increasingly comples. The amount of data explodes with every store, format and number of license partner. This also increases the risk of errors and lets attempted fraud remain undiscovered.<\/p>\n
Our Fraud Detection Software gives labels and artists control over their revenues. Each statement can be checked for plausability within seconds. Red flags indicate irregularities in data. Labels and artists can investigate, if this was only a spike in the data or if this is an error or attemted fraud.<\/p>\n
We develop our Fraud Detection Software, which is currently in beta phase, together with the Technical University Vienna. The basic principle is a Compositional Data Analysis (CoDa), a method that has been used successfully in economics, demographics and biology for a long time. Different streams of data are analysed with complex algorithms to identify their relationship which then become predictable.<\/p>\n
The basis for our Fraud Detection Software is our enormous experience and the amount of data we collected in over ten years in digital music distribution. Within this time we collected a million of data points, that enables us to draw conclusions on the correctness of accounting.<\/p>\n
For example the change in revenues is similar in all stores. If Spotify sales increase, they increase to the same extend on Amazon.\u00a0 Deveation from this behaviour is an irregularity and gets flagged. Another indicator is an increased appearance of certain end numbers. The more irregularities are spoted, the more likely an error or attempted fraud happend.<\/p>\n
At the moment irregularities are detected with 95% certainty. This will improve over time, the more data we analyse.<\/p>\n
The development of our Fraud Detection Software is in full speed. We plan to launch it by the end of 2018.<\/p>\n","protected":false},"excerpt":{"rendered":"
Labels and artists have to trust stores when it comes to collecting their revenues. Error and attempted fraud are hidden inside huge amounts of data. We want to change this.<\/p>\n","protected":false},"author":1,"featured_media":10925,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"image","meta":{"om_disable_all_campaigns":false,"_monsterinsights_skip_tracking":false,"_monsterinsights_sitenote_active":false,"_monsterinsights_sitenote_note":"","_monsterinsights_sitenote_category":0,"footnotes":""},"categories":[13],"tags":[],"class_list":["post-10374","post","type-post","status-publish","format-image","has-post-thumbnail","hentry","category-general","post_format-post-format-image"],"yoast_head":"\n