At Beatdapp, we’re making extraordinary investments in research and development to create the industry leading streaming fraud detection and mitigation platform.
Our algorithms are precision-tuned for the industry to catch both obvious, and not-so-obvious, fraudulent activity, and to avoid false positives. Beatdapp’s dashboard and API provides the context for every flagged activity to easily understand the rationale for each intervention, and our recommended enforcement action.
We’ve recently launched a closed beta for our industry leading streaming fraud detection product. As a proud A2IM Member, we’re sharing with the Indie community that we have space left for only one more DSP to participate in the beta, and we want your help filling it. We know that fraud directly impacts your labels, artists, and marketshare. If you are, or know, a DSP looking for help in the fight against fraud, send them to Beatdapp!
It’s simple: music streaming is expected to reach 1.2 billion paid consumers and generate $75 billion in revenue by 2030. With hundreds of DSPs and connected applications leveraging music worldwide, our industry is now poised to earn the kind of massive, stable, online revenues that make it a prime target for fraud.
In a way it’s sort of a badge of honor that streaming fraud is taking off. Bad actors only spend time and effort on defrauding things that have a high potential payoff. So in that sense, as an industry, we’ve entered a big league alongside digital advertising, e-commerce, banking, and more recently, online education. And yes, those are all verticals we look at for inspiration as many fraudsters transfer strategies and tactics between sectors.
If you believe, as we do, that stream fraud is a persistent issue plaguing streaming platforms around the world, the thought that each one of those services is recreating roughly the same solutions is painfully redundant. We think the path forward is what we’re building at Beatdapp: an independent service that continually invests in R&D to create the industry standard in fraud detection capabilities. We see this as both the best outcome for the industry as a whole, and on a DSP-by-DSP basis.
Here is a sneak peek of one critical screen in the webapp (we haven’t unveiled everything publicly yet). This is a shot of the song level details view on a track (which we’ve anonymized) displaying a high suspected fraud score: 87 out of 99. We score every interaction, highlight the top 5 flags resulting in that score, and define those flags in-product. We’re taking the extra step of including definitions to build a shared vocabulary for what is, and is not, streaming fraud across the industry.
Our product is live with customers, and getting better every day. We have been able to recreate and fully automate existing, manual processes, and discover an incremental ~10% of suspicious and fraudulent activity on platforms that are already devoting internal resources to fraud detection. Deploying our product frees up resources to focus on higher priority, customer facing projects and improves the overall quality of fraud detection. It’s a clear win, and we’re just getting started.