Metrics That Matter — Minutes of Video Transcoded By The Livepeer Network
One of the challenges that open protocols built on blockchain technologies face are how to best demonstrate their usage and adoption. Projects are often compared by abstract concepts like number of transactions, total value locked, amount staked, number of token holders, or even token price — all of which allow arbitrary comparisons between projects, but none of which tells the full story of real adoption. The Livepeer community has always built with practicality in mind, trying to deliver a network which is useful to video developers building streaming applications. As such, we believe that the network should be judged by its adoption amongst these developers, and is best represented by the metric minutes of video transcoded, which represents the number of input minutes of video streamed into the network for processing.
In the above chart, visible at explorer.livepeer.org, users see the weekly buckets of the minutes of video transcoded on Livepeer. In summer of 2020, the network was processing 10,000–20,000 minutes/week. By fall it was routinely exceeding 100,000 minutes/week, and recently it has been surpassing 400,000 minutes/week. This growing adoption should be exciting to not only the Livepeer community, but to the web3 community, as it serves as validation and a strong example of a decentralized work protocol built on Ethereum that is meeting the $70B streaming video market, and solving real problems for real users that exist today.
Much of this growing adoption has been driven by user generated content applications using the livepeer.com API, such as PlayDJ.tv, which leverage Livepeer for its cost effective and scalable video processing for livestreams. Other sources include direct public network integrations like file.video, and performance testing framework that assess the node operators performance on realtime transcoding.
This increased usage has also come with an increase in fees for node operators, who are now seeing about $3000-$4500 worth of fees in ETH paid out each week, on top of the LPT that operators are earning. While fees may be the metric that node operators care most about, minutes streamed represents a more video centric message that is indicative of the actual adoption and usage of the network by its users. Fees may swing due to a number of factors, including the efficiency of the market and price of ETH, but minutes transcoded is a pure metric that indicates Livepeer’s true utility to video platforms.
A half million minutes per week in usage is a great and validating start, but there is still a long way to go to achieve the billions of minutes per week that the network should be capable of handling at scale, due to the video-miner opportunity and capacity of the network. The best thing that the community can do in the meantime is to encourage and support the development of great video applications on top of Livepeer, that seek to empower users and content creators to take control of their content, and establish sustainable economic relationships with their audience and the platforms that they use to stream. Let’s look at and celebrate the milestones along the way — 1M minutes/week, 10M minutes/week, and onwards as we grow — but in the meantime, the Livepeer community can be proud of the up-and-to-the-right adoption curve of the network. While it’s already one of the most utilized work protocols in web3, it can soon be one of the most adopted pieces of infrastructure in video streaming as a whole.