Content Protection: The State of Watermarking

There’s no denying that content security is more important than ever. It’s also hard to refute that the most secure solutions are usually the least simple, impeding workflows and in turn, encouraging workarounds. We know what the best security practices are — watermarking among them — but cost and complication often prevent content creators from effectively implementing them. The amount of content produced may be exploding exponentially, but so is the range of technologies to support that upsurge. This eBook aims to examine the tradeoffs presented by the watermarking options currently available — and why the tradeoff between security and ease of use may no longer be a compromise.

The Case for Security
Whether or not you believe that we are in a content creation bubble, the stakes around content are undeniably higher, and the old adage “Content is King” is making a resounding comeback. As disruptive new OTT content creators (Hulu, Netflix, Amazon) are challenging the traditional broadcast model by creating on-demand access to original content, existing broadcasters — not to mention consumer brands — are trying to catch up with original programming of their own. When that content leaks early to a non-paying audience, it’s a costly consequence.

Feature films that leak before theatrical release can expect to see a 20% hit to box office sales.

Even if some content is “leaked” intentionally to drum up interest, any content that bypasses paywalls, original broadcasts, or subscription systems and winds up unmonetized on torrent sites is bad for the industry. And it’s not just bad for the studios or the networks, it’s bad for the entire ecosystem.

In 2016, every Academy Award nominee for Best Picture appeared on torrent sites prior to release.

The facts certainly will challenge your faith in the good intentions of consumers and in the myth of “if you make content easy to access, they will pay”. But surely we can trust that our peers in the production business value security and will do just about everything to protect content, right?  Unfortunately, that is not always the case either, as evidenced by the leak of on-set stills, rough cuts making the rounds and showing up on Twitter, and entire screeners seemingly swiped from executive desks showing up on torrent sites.

Piracy costs around $20.5 billion annually in the U.S. alone — MPAA.

Content leaks don’t necessarily represent a flaw in human nature. Current industry trends are exacerbating the problem:
An explosion of content. From 2009 to 2015, scripted series orders grew from 209 to 412, or by almost double. That’s not including reality shows, game shows, or feature films.
Strapped staff. We may be making more content, but we aren’t adding staff at the same rate. Our production teams have to do more with the same amount of people. In addition, a trend towards freelancers is increasing the number of people in the supply chain that are not in our immediate sphere of control.
Systems creep. The number of tools and services we are using is growing and as a result we are increasing the number of systems that can fail, fly under the radar of content security and operations, or pose complex workflow issues.
The strain we put on the people who produce our content inadvertently leads to leaks. When deadlines are looming or a team is stretched thin, shortcuts represent the path of least resistance. That means sharing rough cuts without using proper channels (which usually require extra time or work).

So how do we avoid or at least minimize shortcuts? While creating a culture that values security and accountability is extremely important (that’s a topic for another eBook), so is selecting the right technology and tools for the job.

There is a trope in software design called the Security Triangle. It explains that the three objectives in system design - Usability, Cost, and Security - are difficult or impossible to achieve at the same time. In fact, Security and Usability are typically antagonistic.  Consider the recent change in ATM cards. The security chip makes the transaction safer, but the user experience arguably suffers because the process takes longer, requiring the consumer to keep the card in a machine, AND still demands an electronic signature.  

Choosing technologies that mitigate the tradeoff between security and usability is key for anyone putting systems in place for the media and entertainment industry. While watermarking is regarded as an important tool in the security arsenal, it may just be the silver bullet.

Watermarking - The Overlooked Silver Bullet
Let’s take a closer look at one security technology that’s being radically changed by the power of the cloud: Watermarking. It’s generally considered a best practice, and there are several methods currently available, but each represents a compromise between security, ease of use and cost. Watermarking is a loose term for media that is marked in a specific way in order to prevent theft and track down the source of origin. The common forms of watermarking today are: 
Forensic / Digital watermarks
Visible watermarks, also called burned in watermarks

Forensic Watermarking

Perhaps the most common form of watermarking, Forensic Watermarking, has its roots in distribution to cinemas. It involves placing an imperceptible mark within the code of the video.  Its early use in theatrical distribution ensured that the user experience was unchanged but content was traceable. If a movie theater leaked content either by way of a guest recording off the screen or unauthorized digital reproduction, the content could be tracked to the theater. The same principle applies today for digital files.


  • Imperceptible to the user


  • Its strength is its weakness: since the watermark is invisible it does not act as an effective deterrent
  • Minimum content length requirement. Not applicable for cuts under 2 minutes
  • Expensive
  • Difficult to scale and embed into production/post-production workflows
  • Does not deter from screen-scraping unless used in conjunction with a visible watermark

Visible Watermarking
Visible watermarks are easy to spot as they commonly contain a disclaimer such as “Property of Acme. Not for distribution”. In some cases the name of the recipient is burned in. Considered a very effective deterrent, the process of burning in a visible watermark is time consuming.


  • Effective deterrent, especially if the name of the recipient is burned in


  • Some viewers consider it distracting
  • It’s time consuming and can take up to 75% of the duration of the original piece of content to complete
  • Costly to scale and embed in most workflows

Overlays became popular with the emergence of online video and are still used frequently. The underlying video stream is raw and unaltered and the watermark is overlayed by the player.


  • Inexpensive, as the video itself does not need to be altered
  • Acts as a deterrent since it can display user identifiable information
  • Can be used easily in most workflows


  • Completely insecure as the video stream is left untouched and can be stolen

Although DRM is not technically the same as watermarking, we often see DRM mentioned in the same context. DRM systems are primarily concerned with the secure delivery of a key that can be used by a player to decrypt content. Often, the player has options for displaying overlay watermarks.


  • Inexpensive, as the video itself does not need to be altered


  • Does not deter from screen-scraping unless used in conjunction with a visible watermark
  • DRM landscape is complicated with multiple solutions required to satisfy multiple platforms and devices
  • User is required to install software before content can be played back. This can be an issue, especially if users do not have administrative rights

None of the conventional options examined above check off all three objectives in the Security Triangle: security, usability, and cost.

However, a new breed of watermarking is now available that combines the best aspects of visible watermarking, without the cost and resource constraints. SafeStream is an on-demand watermarking solution that embeds user identifiable information as a visual watermark into video files in real-time, using massive parallelization in the cloud.

Here’s how it works: Every time a user wants to view content in a browser or on a mobile device, a request is sent to watermark the video on demand. This means a 90-minute video with a personally identifiable watermark can be available to the viewer in as little as 10 seconds. It also means that watermarking is now fast and cheap enough to incorporate into multiple workflows — from dailies and rough cuts, through legal review, dubbing and executive viewing, all the way to screening of final pre-release content.


  • From short clips to feature length movies, video files are uniquely watermarked in seconds
  • Does not require proprietary player software which means it works out of the box on mobile devices and in web browsers
  • Visual deterrent that clearly identifies users by their name, email, subscriber id, or IP address. Various options exist to customize the watermark placement, duration, and look
  • Cost-effective, as files are only watermarked as they are requested
  • Can be implemented seamlessly into existing workflows, such as media management systems, screening rooms or Apple TV Channels

Want to learn more about SafeStream?

Schedule a demo today