A recurring nightmare for marketers and PRs is the leak of a product before its official launch. Yet they could avoid it if they knew about steganography. This technique is well known to forensics and security experts for hiding information in images. Here's how to use it to secure embargoed visuals and detect potential leaks.
Quick read :
- What is steganography?
- How steganography improves the security of images and videos
- Why protect your product campaign visuals with steganography?
- Which marketing materials need more protection?
- When and how to apply steganography to your campaign visuals?
- Can leakers remove a steganographic tracker?
What is steganography?
Steganography is the practice of concealing a file, message, image, or video within another file, message, image, or video. It is sometimes associated, wrongly, with hacking activities as being a means of "hiding bad things in good things". It is just the practice of hiding secret messages in otherwise non-secret mediums.
When the non-secret medium is an image (or a video), the most common method is to modify the value of pixels so that the human eye cannot detect changes in the image or video frame. This technique allowing to embed a secret message within an image is also called digital watermarking.
The term watermarking is sometimes confusing because it reminds of visible watermarks used to sign an image. However, the name Digital Watermarking is more used than Image Steganography.
How steganography improves the security of images and videos
In our article about the different uses of digital watermarking, you can learn that steganography has long been used to verify the integrity of content, to detect illegal copies, to identify their perpetrator, most often in the media industry.
The code (or payload) concealed in the image acts as a marker or invisible identifier. The embedded data can be of any kind but is usually in the form of a text or alphanumeric ID. Its length will be limited by different constraints such as the size of the output file, the perception level of the image alteration, the expected robustness to further modifications (compression, crop, resize).
The purpose of the invisible identifier is to remain hidden in the image or video file, regardless of the changes or uses made of it. One should still be able to detect the invisible code's presence and read its content, even if it appears in a printed magazine or a TV show, cropped or slightly rotated.
Furthermore, imagine that the steganographic information contains the coordinates of the image's initial dedicated recipient, like a social media or email address. Then you would know more about how this image leaked and improve your security procedures accordingly.
Why protect your product campaign visuals with steganography ?
As detailed in our article on product leaks and how they happen, there is a lot of marketing and communication material that needs to be shared before the official launch with resellers, influencers, marketplaces, journalists, public relations agencies …
A brand must thus strictly enforce its embargo period before launch. From blind trust to legally signed NDA, no deal is 100% guaranteed to be honored. Most of all, when a leak does occur, one needs to know where it came from to take action.
This is where steganography comes in.
Even after a launch. Brands also need to monitor their visual content to assess a campaign's performance. Visual content analytics is a smart way to enhance traditional web analytics, but be careful : images found by simple reverse search (like Google Image) need to be certified as yours, and not just similar content or false positive. This is where, again, steganography (or digital watermarking) plays a significant role. If the code is found in reverse search results, then it 100% confirms the source of the image. It is particularly useful for product images with simple shapes or with few distinctive details, like packshots, tires, or electronic parts.
Which marketing materials need more protection ?
Marketing materials are the top origin of leaks. Pro leakers who hunt them are "merciless", as mentioned in this article about Leaked Galaxy S20 marketing materials. They seem to forget that "stripping Samsung's execs from any hope of surprising us with anything" is not a game but a severe loss of income.
In many industries like automotive or consumer electronics, leaks on new models drive down sales of models still on the shelves. People stop buying, waiting for the latest model to be available. It has been the case for Apple with an unexpected iPhone sales fall on leaks about the new model. Counterfeiting is the most prevalent use of product leakage of visuals in other industries like apparel and luxury goods industries.
For all these industries, most visual assets should be watermarked each time they are delivered, internally, or outside the organization, for sharing or just contributing to the workflow.
- Lab prototypes
- Project mockups
- Product designs
- Fashion studio shootings
- Events preparation documents (slides, speech drafts)
- Embargoed press releases and press kits
- Print marketing materials (before sending to the printer) like posters, billboards, goodies, brochures …
- Marketing research content
It includes videos, not only promotional but also internal videos, like sales training and tutorials.
When and how to apply steganography to your product campaign ?
There are two ways to lose sight of your content : when you send it to someone, and when you allow access to it (whether protected by DRM, downloaded using login/pwd, right-click "save as", screenshot from display …).
That's why the perfect time to mark your marketing content, even in progress, is BEFORE it leaves your hands. Before sending copies to several recipients. Before publishing on your website, your PIM, your DAM, your press room, your social media.
Tools such as Imatag Leaks allow you to batch-watermark your campaigns using proprietary technology. It is a real time saver compared to mainstream digital watermarking solutions. Moreover, be careful, some of them cannot apply different watermarks per copy of a single image.
- Upload your assets, register your recipients, the online service generates a unique watermarked copy per recipient. Then you just need to download your files and send them away.
- If you spot a suspicious copy somewhere online, you just need to upload it (or a screenshot). By reading the watermark imprint, Imatag will reveal who the original source of the leak is.
Can leakers remove a steganographic protection ?
To make the source image identifier unreadable, a leaker would need to modify or delete the pixels embedding the code. This means that if the pixels are located in a single area, like the wallpaper of a smartphone render, replacing it will remove the code. This is what happened in the following example :
Another way to cover tracks : leakers compress the images, sign it with their own visible watermark, to modify as much as possible the pixels carrying information.
But not to worry, this has no impact if you have the right steganographic software. It must rely on a professional digital watermarking technology, at minimum complying with the following selection criteria :
- The watermark must be robust enough to withstand compression and resizing
- It must be imperceptible
- It must cover all areas of your image
- Without compromising its quality
Test IMATAG watermark here to check that your product images will keep their quality even compressed or modified, and still allow you to certify the source of a leak.