With 500 hours of content uploaded to YouTube every minute, it is impossible for humans alone to keep up with the various nuances of brand suitability of each brand. When you don’t get it right, the toll can be great:
- Wasted dollars: 52% of videos made for kids on YouTube are not tagged as such, causing brands to waste money on non-converting audiences
- Damaged brand perception: 50% of consumers say their view of an advertiser is negatively impacted when an ad appears alongside undesirable
Most – if not all – brand suitability solutions deliver some form of protection for brands; however, only those specifically taking advantage of data science deliver the balance of protection and reach. A big risk of not taking a data science-driven approach is that you will block good supply due to erroneous categorization and miss the opportunity to reach your audience. Additionally, this missed opportunity can be compounded by the risk of letting not only unsuitable – but unsafe – content through your filters. The best brand suitability solutions are built to understand the nuances of context. The ability to interpret contextual nuances at scale is what differentiates data-science-driven solutions.
One example of this is keyword blocking, a widely accepted way to approach brand suitability. However, when relied on solely as a tactic, keyword blocking is inefficient. Keyword blocking looks for a specific word in a message and does not take the context of the message into account. We’ve seen other third parties who over-emphasize keyword blocking have ~75% accuracy using keyword matching as their approach. Is that good enough for your brand?
Take the word “shot”, for instance. Using keyword blocking alone, we found that the word “Shot” blocked 736 videos and 172M views on ESPN alone. However, by taking an AI-driven approach, machine learning analyzes thousands of videos to determine the context around the word “Shot” and understand when it was used in a sports-related way to avoid overblocking content erroneously.
Keyword Matching vs. AI-Driven Brand Suitability Scoring
AI-Driven Brand Suitability enables you to be part of the right story without overblocking.
Innovation that Helps Advance the Solution
Advances in data science have had a positive effect on brand suitability solutions. At Pixability, we combine generative AI with human review to deliver 99% brand suitability for our customers.
BrandShield, our patented YouTube Brand Suitability technology, combines data from thousands of campaigns and hundreds of millions of YouTube videos and channels to ensure you’re executing brand suitable campaigns – while still driving performance at scale. Additionally, our team leverages ChatGPT to drive 100X scale in our training data to deliver more precise Machine Learning scoring.
Ready to advance how you do brand suitability? Reach out to Pixability today!