Viewpoint for menu pin class add

Unveiling the Blindspot: A case study in identifying brands, products, and services in social media videos


In the vast world of social media, video content has rapidly become a cornerstone of marketing strategies. However, extracting valuable insights from these videos can often feel like trying to see into a blindspot. With evolving speech-to-text technologies and the ever-growing popularity of platforms like Twitch, Instagram, TikTok, and YouTube, the task of identifying brands, products, and services in videos has never been more crucial, yet equally challenging.

The Current Reality

The ability to extract value from video content significantly relies on speech-to-text technologies. These tools, powered by artificial intelligence, are rapidly evolving and becoming increasingly accessible. However, while this technology can provide transcripts or closed captions, the true essence of the content often gets lost in translation, turning video substance into a blindspot for marketers.

Navigating the Marketing Maze

As marketers, it is inherent in our roles to stay informed about the operations of our campaigns. Whether it's a commercial relationship through sponsorships, pay-by-product, or affiliate commissions, understanding the content produced, its timing, and its adherence to jurisdictional requirements is paramount. This is even before we consider contractual obligations, product representation accuracy, and success measurements.

The Scale of the Challenge

The task of identifying all involved work products is hindered by a multi-faceted problem. The simple act of booking a creator can lead to a multitude of assets across various platforms. The complexity of tracking the reach and impact of these assets, coupled with ensuring legal and business checks, can be overwhelming. Furthermore, the limitations of commercially available transcription services can make identifying brands, products, or services a daunting task. As RippleXn CEO, James Beaven explains, "If you can't accurately identify the brands, products or services talked about in video, any insights or analytics that follow are significantly devalued."

A beauty #GRWM (Get Ready With Me) YouTube example

Note: This video was taken at random from a wider research piece. RippleXn has no affiliation or relationship with the creator, nor currently any brand or agency involved. 

Sam Chapman's YouTube channel has nearly 42k subscribers. Her video from the 11th of January 2024 at the time of this article has >32k views. 

Watch it directly on YouTube and review the transcript yourself.

The following 6-minute video serves as an excellent illustration of the limitations of off-the-shelf solutions when it comes to identifying brands, products, and product variations, particularly those with unusual spellings or phonetic considerations. 

Here are some brand examples. Look at 26 seconds in. She is talking about a Shiseido skin care product.

YouTube’s own transcript reads:

“I’m onto my Cho Skincare”
“this is the old Aline um line”

If we look at the RippleXn’s video report, we can clearly see the correct identification of Shiseido skin care along with NuFace, Maybelline, Clarins, Glossier, Murad, Tom Ford, Victoria Beckham, Guru Emporium, NYX and many more. 

In the video, RippleXn correctly identified 16 brands, products and beauty accessories. If you were to rely on hashtags, video tags and written descriptions you would only get the below.

This is why the current text based tools are not fit for purpose in the video dominated world.

Why This Matters So Much

The example of Sam’s video isn’t a criticism of YouTube’s transcription process - it’s a hard thing to do at scale.  But it is a really good example to show how much the substance of the video, the spoken word, is currently still a blind spot.

Imagine being able to use this to accurately measure your campaigns. Manage and de-risk your liabilities against advertising law and extra legislation like financial services selling or age-restricted products.

But also consider research. Finding out about how your product is used, used in combination with other products. Which demographics are using, talking, and raving about your product. Imagine monitoring beauty vloggers over multiple platforms and being able to compare your product against a direct competitor. This would be an ESoV metrics you have never been able to have before.

What Can I Do Next?

If knowing more is important to you, or you would like your company’s products researched, talk to the RippleXn team at hello@ripplexn.com or via the onsite forms.

 

 

 

 

Back to blog posts

Top