14 Jul Perspective | XL8: Driving Affordable Localization for the FAST Channel Market
AI-based media supply chain companies bring to market new ways to globally monetize
FAST content through cost-effective, automated localization
July 14, 2023 | Global content distribution veterans have all “been there, done that” when it comes to localizing media content for broadcast, cable, DVD and Blu-ray, and all the VOD services. Making decisions on what content to localize in which territory, and in which language(s), was pretty simple and clear cut: You sold or licensed a piece of content to a company or broker in a foreign territory for a specific license fee and a license period. You knew exactly the dollars you would get from the license fee, and from there you could easily plan your ROI in terms of your localization costs.
Sell a title into Germany for $1 million for a three-year license fee, no problem. You’ll agree to spend $30,000 on localizing that content for the territory. In many cases, localization costs could be rolled into the fee paid by the licensee in the territory, thereby creating a no-cash out-of-pocket scenario for the content’s localization needs.
Paradigm shift for greater revenue potential
The speedy rise of FAST, however, has flipped the paradigm 180 degrees. The good news is that FAST channels can create a major upside in licensing revenue, much of it incremental for previously exploited content, and new paths for new-release or user-generated content that might not otherwise find a home on the major SVOD platforms. But the FAST world also comes with a greater degree of risk and barrier to entry for global content.
Unlike traditional licensing models, the majority of the FAST platforms pay license fees through a revenue sharing model. The owner of the content takes a negotiated share of the ad revenue as payment for licensing the content to that FAST platform or FAST content aggregator. The problem this creates for content owners is that the FAST platforms typically wait up to three months or more after the program has aired (er, streamed…) to calculate and pay the shared revenue.
But revenue sharing makes ROI hard to predict
This now creates a new problem when it comes to localizing content for foreign language territories (or foreign content into English), as you are now “flying blind” when it comes to trying to calculate your ROI for localizing that content, especially if you don’t get paid for it until more than three months after release. Further, you have no idea or guarantee that the content will perform well enough to cover your localization costs, especially when utilizing pricy traditional, manual localization processes. For previously released content, you may be able to recover the existing localized files, but they may be unavailable or unusable, especially older catalog content.
The solution: As the FAST channel market embraces advanced streaming and advertising technology, AI-based media supply chain companies have brought to market new ways to globally monetize content through cost-effective, automated localization processes. This includes a plethora of AI-driven functions along the media supply chain, such as generating automated timed text (for re-creating scripts), subtitle generation, machine translation and localization, synthesized dubbing, ad-insertion points, content compliance, and much more. Some companies are even putting all these services together in one streamlined package.
Imagine the following scenario: You have a 20-year-old movie that’s two hours in length and was originally created in English. You license this movie to a FAST platform for French-speaking regions around the world. You know from previous experience that localizing this content just for subtitling and dubbing typically runs about $20,000-to-$25,000 for two hours of catalog-type content (much more for new release “tent-pole” type content). What you don’t know is how much you’ll be paid at the end of the first three-month period, nor what your projected ROI will be. So, how do you justify moving forward with the localization process? That’s where AI and new technology fit in.
AI can bring down localization costs
In this same scenario, you could easily create two hours of subtitled content for under $500 using only the power of AI machine translation and auto-subtitling built into the platform. For most top-tier language pairs, including English to French, the best AI translation companies that specialize in media and entertainment content are hitting upwards of 90 percent accuracy, and in some language pairs over 95 percent accuracy. Add in another couple of thousand dollars for basic synthetic dubbing quality, and now you have new decisions to make.
For under $3,000 with dubbing, or under $500 with just subtitling, you can now localize and monetize content, with commercially acceptable quality (for many types of content) without betting the farm on your ROI. Want a little higher quality level? Many of the top language service providers (LSPs) have realized the importance of the FAST market and are quickly creating hybrid models that utilize a lot of the new AI functionalities, but then add in an element of human “finishing touches” that create an end result closer to the 100 percent quality realized by traditional localization methods.
Finally, it’s time to ask yourself the $25,000 question: For my specific content type, is it worth it to spend $25,000 to localize my two hours of content from English to French and get 100 percent perfection, but not know if I’ll recoup it? Or, to get a better handle on a new market, or further penetrate an existing foreign market, am I better off spending a quarter of that for a quality level closer to 90 percent to 95 percent and take the ROI risk? I suspect the answer is the latter for lots of non-monetized content that’s out there. Even better, computers don’t complain about one language being more expensive than another, or whether linguists are available for a tight turn-time. To them, it’s just computing power. And in fact, if you are distributing content to a region with multiple languages, the computers can translate all the languages at the same time, in the same pass, thereby saving even more time and money.
At the end of the day, these new technologies bring important new options to global content monetization and help take down the barriers to distributing worthy programming to new audiences. Not everyone drives a hand-built luxury car. A Toyota will get you from point A to point B in the same time as a Mercedes, but at least you have the choice to drive what suits your taste and budget. Now, for driving affordable localization for the FAST Channel Market, you finally have that choice as well.