System Federation and Abstraction is the only path forward for Media Companies
Coming from the on-premise, SaaS, and system integration worlds in video technology — the one sales mantra that I recall with the most clarity is “solution selling” and “system lock-in”. The reason why Wall St (and Sand Hill Road) love SaaS so much is the defensibility of a product that’s difficult for a customer to switch out of. That business strength on the vendor side quickly translates into complacency and thus reduces the likelihood for product innovation for the client side.
Having spent some time on the client side of the media world with VICE and Abu Dhabi Media — its increasingly clear that what’s great for the vendor industry does not create a good foundation for a fast moving moving media-based enterprise. Media companies are often pitched on a “total solution”, and while that’s highly convenient for purchasing managers and CAPEX planning, it stunts product development pace and institutional understanding of the fundamental technologies upon which media travels.
I’ve long maintained that media companies must do their best to innovate the medium — in this post I’ll outline some of the guiding principals of how to go about doing that.
Media portals are not products — they are glued collections
Media portals (be they VOD or Live stream-based) aren’t “products” — they are collections of products, services, and 3rd party APIs wrapped as a “solution” and delivered to a client with an often tightly tied bow. In an effort to effectively get to market, content companies are often forced to tie their content transformation and transport mechanisms (ingest, transcode, CDN, etc) to their business logic (meta data, user management, ad management, etc) systems which means creating new consumer-facing properties only solidifies the lock-in.
To be fair to the vendors — this is not entirely the industry’s fault. In an attempt to grab investor coveted market share, business logic systems are frequently undervalued while transformation and transport systems are re-sold with a margin. At scale, paying for an online video platform is basically paying for their margin on Amazon and Akamai.
With that hard truth established, how can content companies stay nimble in distribution while building a foundation for a wide array of content monetization businesses?
The answer is not trying to squeeze the vendor margin — it is investing that margin in internal knowledge to ensure that mechanisms for content management and distribution do not become anchors that slow the organization’s ability to pilot and release forward thinking products.
The key to this approach is twofold: internal expert steering teams as well as complete federation of functionality between the technological building blocks (or the “stack” if you will) that manifest themselves in various human consumable interfaces — for both internal administration and external consumption.
What is federation of functionality?
Simply put, federation is proactive management of sub-systems to minimize interdependencies of vendors. If there is a new transcoding engine, an updated player technology, or a more modern method of cataloging content — an organization should be able to swap out specific functionality in an elegant manner without a massive costly infrastructure project that takes ages to implement. The diagram below is a crude way of thinking about federation, but should get a conversation started at your organization about what parts are best of breed in the markets versus what’s bundled in by your current platform or system integrator of choice.
Obviously this is not exhaustive. Some other abstracted systems to think about are:
- written and image content (via content-only API)
- access control and username management
- content distribution restriction (biz engine of geo rights)
- packaging and delivery engine (transcode, format wrapping)
- analytics systems (event data warehousing and display)
- authoring environments for written content
- publication control (permission, calendaring)
- front-end user interface technologies
- AWS glacier-type archiving and backup of core media catalogues
In the video world, some good vendors to look at for this functionality include Uplynk (from Verizon), Glacier (from Amazon), Segment (for event data), JW (for player management), Vigour (device-sensitive UX framework), Contentful (cms-less content API) as well as plenty of others that will be a fit for your specific organization.
What is the role of internal expert groups?
Living in the “Chief Digital Office”, this should be your ultimate source of technology truth. The folks here need to have specific and deep subject matter expertise (software architecture, media processing, mobile engineering, product design, etc) and ultimate authority to make business decisions to relieve organizations of vendor lock-in.
One of the pleasure of using Segment, for example, is the ability to upgrade your organizations ability to parse data — going from simple solutions like free Google Analytics to deep SQL analysis tools like MODE as your organization evolves.
How to get started?
For a new organization, getting an internal product steering team on board in the beginning is key. For existing players though, finding a budget for these bodies should be done through an analysis in current systems.
Even the most sophisticated organizations (save for fully managed internal teams like Netflix, Amazon Video, etc) have costs associated with “vendor margin leakage” where the company is getting something resold to them. That re-sale margin (often in the tens of thousands of dollars per month) is the place to start.
If you are a broadcaster, aggregator, telco, or other organization where video software engineering is not at the core of your business — I highly suggest taking the first steps in adopting an approach that will assure that your digital market presence does not stagnate.
I always enjoy jamming with media folks that would like to learn more. Say hello!
If you enjoyed this article, please favorite and share ❤
Thanks to everyone who helped edit!
Much love. G.