The Audience Targeting Mistake That’s Costing You Customers
Sean Black, GM of AudiencePath, the real-time audience company from Dailymotion, on why audience targeting is broken and what’s working now.
Most marketing reaches the wrong people because it targets what audiences did yesterday, not what they are doing right now. In this episode of Liftoff with Jeanniey Walden, Sean Black, GM of AudiencePath (the audience company built on Dailymotion), explains why audience targeting is broken and how real-time signals from more than 5 billion monthly video streams help brands reach the right person in the moment, train AI on data they actually own, and tell AI content from real content.
What you’ll learn
- Why audience targeting is broken: most brands target on stale data instead of real-time intent.
- The shift that works now: reach people based on what they are doing today, not last month.
- Why owned data wins: AudiencePath trains its AI on its own signal from 5 billion-plus monthly streams.
- Intent beats format: it does not matter whether someone watches AI or human-made content if the intent is real.
- The next frontier is metadata: identifying content frame by frame.
The conversation
Sean Black: Fundamentally, marketing has been broken. The way people think about audiences and creating audiences has been a challenge.
Jeanniey Walden: Sean, welcome to Liftoff. It is so good to see you. We don’t usually hug on the show, but this was special and we needed to do it. You are doing something so many brands should be doing and so many agencies should be thinking about, but nobody is thinking about it yet. You are using audience behavior and identity in a unique way. Tell us all about what you are doing.
Sean Black: What’s exciting is that fundamentally marketing has been broken. The way people think about creating audiences has been a challenge. We are a company that owns technology, video player technology, so we see over 5 billion streams a month. Using that stream lets us create signals. We believe in using the right signals to identify people, not by what they did yesterday or a month ago, but by what they are doing today, so you can reach them in the moment you need to.
Jeanniey Walden: We have so much happening in 2026 with new technology, AI, and new ways to engage. Why do you think so many companies are still using old, traditional ways of doing media targeting?
Sean Black: First, it costs a lot. It takes real cost to build the agentic platforms that companies and agencies are trying to create. But the biggest part is training. You can’t just spin up an AI platform and expect it to do what you need. You have to train it with the right data and the right information. That is what we are excited about: we own the information we feed it, so all of our models are trained on our own data.
Jeanniey Walden: With AI creating so much content, and a sea of sameness in video, how do you see the landscape changing? How do you discern whether someone is watching AI content or real content, and whether to recommend a brand make AI content?
Sean Black: Great question, and it comes back to who Dailymotion is. At the end of the day we are a video company. We are an SSP with a video player on over 5,000 publishers. Our tech is trained to better understand AI-driven content or natural user content. I am not sure it matters what the end user is watching if the intent is there. If they have the intent and the interest we are looking for, I don’t care if you are watching an AI-created automotive spot, because it clearly is something you are interested in. The biggest challenge is going to be the metadata. How do people distinguish AI from real content? Our video player watches frame by frame and creates metadata from that video, so we help publishers and users have better identified content.
Jeanniey Walden: I love that. That’s super smart. This was an amazing conversation, Sean. Thank you so much for being on Liftoff.
Sean Black: This is great, thank you.
Jeanniey Walden: And that’s a wrap for today’s episode of Liftoff. I am inspired. Follow us at Liftoff Show on every social channel, and make sure you catch our next episode.
FAQ
Why is traditional audience targeting broken?
It relies on what people did yesterday or a month ago instead of what they are doing right now. Real intent lives in the moment, so stale data reaches the wrong people and wastes spend.
How does AudiencePath target audiences differently?
AudiencePath, built on Dailymotion’s video player technology, sees more than 5 billion streams a month and turns that real-time stream into signals to reach people in the moment, training its AI on its own data rather than borrowed data.
Why do companies still use old targeting methods?
Two reasons: cost, because building agentic platforms is expensive, and training, because AI needs the right data to work. AudiencePath owns the data feeding its models, which is the advantage.
Does it matter if people watch AI-generated or real content?
For targeting, not really. What matters is intent and interest. If the intent is real, the format of the content is secondary. The harder problem is the metadata that distinguishes AI from real content.
How do you tell AI content from real content?
The video player watches frame by frame and creates metadata from the video, which helps publishers and users identify what is AI-made and what is natural content.
Put this to work
Want more of what’s actually working? Get the next playbook in your inbox.
Subscribe to the newsletter Subscribe on YouTube · Apple Podcasts · Spotify · The Trust Threshold (book)