Organization
Pearson
Industry
Education
Headquarters
London, England
Solutions Manager
Kyrce Swenson
Teams Using the DAM
Content Production, Marketing
Delivering materials that help the world learn
With more than 160 million users in over 70 countries, Pearson is the world’s largest learning company. It’s also the fourth largest publishing company in the world, providing valuable tools for teachers and students.
But as the world went digital, Pearson recognized their user base preferred learning materials in video format. To accommodate them, Pearson developed its own proprietary video player with millions of video assets users can access. A typical month sees an average of 70 million video plays, with those numbers rising during the busy back-to-school season. But with internal assets scattered, it was difficult for the education company to make the most of what they had.
The goal
With no central repository, Pearson’s massive number of videos were under-used. Pearson’s assets were stored on an old digital asset management (DAM) platform that wasn’t capable of handling their current demand. In addition, there were assets stored on physical servers, some located in the U.S. and Europe, which was a costly hosting endeavor for Pearson.
Pearson needed a system that could make their material available to learners at scale almost instantly, with the ultimate goal of meeting the demand of up to 40,000 API calls a minute while also reigning in their hosting costs.
As Kyrce Swenson, the Solutions Manager at Pearson put it, “What I wanted was a way of making sure that we could deliver targeted and enriched video content, and that we could deliver things that would increase outcomes for the people who trust us with their education.”
Pearson needed a solution that would include support for five target areas involved in the end-to-end video process, including storage, transcoding, API services, delivery, and a connection to the Pearson video player.
While providing materials to learners was the key objective, the team also wanted to migrate to a solution that would make it easier for internal users to find the assets they needed. Currently all assets at Pearson were kept within a single business division. Swenson and her team wanted to make their new DAM software accessible to the entire company so all Pearson staff could make use of their incredible library of materials.
Pearson needed a more effective solution — one that would let internal users find what they needed, while cutting the costs incurred by heavy external video usage.
The solution
After searching for a Media Asset Management system that could handle their huge collection of video assets, Pearson selected Orange MAM as their digital asset and media asset management system. Some of the reasons for the choice included:
Connecting MAM and Pearson player
The Pearson system hosts educational videos along with associated files, such as subtitles, audio language tracks, XML files, HLS file formats for mobile.
With Orange Logic, Pearson set up a system that allows external users to access this MAM content without even realizing it. End users call up a video using the Pearson player app. This triggers an API call that the user is completely unaware of.
The specific API call developed by Orange Logic minimizes the number of transactions between the MAM and Pearson’s system. It creates a cached, CDN layer between Orange MAM and the player, so only new videos have to be pulled from the MAM system, and videos that are repeatedly viewed are held in the cached layer. This reduces costs and latency when users call up a file.
Pearson also switched to their own account with Cloudfront (Amazon’s Content Delivery Network), where a volume-based discount will further cut the costs of its massive data needs.
“We've done a lot of optimization and Orange Logic's been really great about optimizing the servers on that,” Kyrce said.
Searchability
Internally, the system was made more usable through the smart application of metadata. For the first time, general users were able to find material housed in Pearson’s collection — without knowing the publication year or ISBN of the asset they needed.
Pearson was also able to apply AI metadata tags to their assets, making millions of assets easily discoverable. Auto-captioning was applied to the content that didn’t already have captions. Those auto-captions were applied as metadata to the assets, making them searchable and saving countless hours in manually generating the captions. For instance, Kyrce mentioned searching for a video about Romeo and Juliet that didn’t have the name of the play in the asset title. She was still quickly able to find the exact asset she was looking for.
By using Orange Logic’s robust metadata tools, Pearson was able to set up a consistent metadata schema and batch edit metadatawithin the UI, and by uploading metadata CSV files.
Uploads are formatted so that assets and their related files have similar file names. Then those files are automatically grouped together so that a video’s caption tracks or associated files will become related to the video without extra work on the part of the Pearson staff.