SPARK Neuro: The Company That Literally Picks Your Brain
SPARK Neuro is one of those companies that makes you ask, “You can DO that?”
They’re neuroscience. They’re advertising. They’re biometric research. And they’re offering an alternative to the way we measure audience engagement. In traditional audience engagement, a subject might tell you how they like a movie trailer by telling you what they thought or by responding to a survey.
SPARK Neuro goes several steps further; surveys are unnecessary because the company can measure audience engagement by reading brain and nervous system activity directly. It has the ability to quantify attention and emotional levels with a precision that’s down to the second.
Some of their customers include Toyota, NBC, Telemundo, and Netflix. They’ve raised $13.5M in a Series A led by Thiel Capital, as well as Michael Eisner and Will Smith (yes, that Will Smith) as key investors.
The company was founded in 2016 by Spencer Gerrol (also CEO), who I recently had the chance to speak with about the mind-reading company.
What's your background? How do you get involved in something like this?
It starts out during my academic career where I was studying human factors. Now, no one's heard of human factors. It's essentially an applied form of cognitive psychology, and so all the way back to what I studied, I've been fascinated by how human beings process the world around them. That includes everything from visual perception to the ergonomics of the chair I'm sitting in right now, so it's all about studying human beings and psychology and applying it.
So I was really fortunate because I got to study my passion. That's when it all began: my obsession with understanding human beings, how we interact with the world around us, and what is the real science behind that in a way that we can apply. My career has really been dictated around that.
And how was SPARK Neuro founded?
In 2007, I went on my first date with a woman who's now my wife. And she asked me what I wanted to do with my life—what I wanted to be when I grew up. And I said, well, I've been involved in research for my academic background, and I noticed one thing. The most important of all the things that we need to understand about humanity are around attention and emotion. If you're not paying attention to me right now, then what hope do we have of creating a great article together? If you're not emotionally engaged with what I'm saying, then the readers will be bored too, and yet there wasn't a way of measuring that.
People do surveys, they do focus groups, and that's a self-reported way of asking people how they feel, but the problem is that feelings are largely subconscious. They're driving all of our actions and all of our behaviors, but there wasn't really a way to measure quantitatively.
And I knew, given my background, that the answer to that question would be to tap right into the source, to measure people's brain activities, and to truly measure in real time second-by-second just as I'm talking right now. When are you paying attention? When are you hanging on my word? When am I losing you? when are you feeling an extreme emotion?
Unfortunately, in 2007, the technology, computing power, and things like machine learning were nowhere near where they needed to be, and so I kept this idea and built my first company, which was like me learning lessons of entrepreneurship and applying psychology to real-world applications for web design.
And then in 2013, as my other company was booming quite well, I saw the technology start to get there, and so in 2013, I hired my first neuroscientist and said, okay, let's work on building these algorithms to quantify emotion and attention, and we spent the next two years building those algorithms and finding the most scientifically valid, academically rigorous way to measure the human experience.
So when I thought the market was ready and when I saw that the technology was ready, we went for it, so at the end of 2016, we were getting a lot of attention. We predicted that Trump would win the election based on a bunch of research that we have been doing in key swing states like Pennsylvania and Florida, and it was catching on and clients were starting to wonder more about it. I exited my other business and launched SPARK Neuro on January 1, 2017. And then it was just perfect timing. The market was right, the tech was ready, and within 8 months we were incredibly profitable. We had something the market was into, and so we thought about scale.
How does the technology work?
The key thing is that the science really needs to be holistic, and there are multiple different factors that are critical to having an accurate understanding of the human experience, and among those sensors, there's one particular one which is by far the best one, which is your brain. The way your brain processes information and emotion are through neurons that are firing off in your brain. And there are small amounts of electricity being released through your scalp.
EEG [Electroencephalogram] is a way to measure the electricity that's emitted from your brain by putting a device on someone's head. The challenge of EEG is that the data in and of itself is incredibly difficult to interpret because you have all these broad waveforms of different frequencies of electrical amplitudes, and all these different locations. Our job is to make that tornado of data make sense, where we can say okay, we have a metric, it goes from 0-10, and we can measure the intensity of emotion.
And at the same time we have other metrics, so we use other metrics like galvanic skin response, GSR, which measures electrothermal activity, which is a fancy way of saying when you get emotionally excited, your palms sweat a little. When you get nervous or excited you get a little bit of that sweaty palms. You don't notice it most of the time because it's so microscopic, but it conducts electricity and that's something you can measure. Simultaneously, we’re measuring your eye movement using eye-tracking technology to understand where you're looking and what patterns you are exhibiting. We’re also using a camera for your facial expressions. While facial expressions don't do much, in combination with these other metrics, it does a whole lot.
How is that data used?
If you're producing a movie trailer, an ad, or a pilot TV show, the things that are going to most predict whether that's successful are whether people are paying attention to it and whether they are are they having an emotional reaction. And so we deliver graphs that say, hey, they pay attention here which is great because it's a critical moment, but you start to lose people here, and so we need to change something.
They're getting the second-by-second data, so imagine the movie trailer that's two-and-a-half minutes. We're saying okay, people are really excited in the beginning, they got bored for the next 14 seconds, you are kind of able to get them back and experience this emotion here, but it was sort of a letdown after that because they didn't then have the payoff, and so we're giving them the granular line graph showing emotion and attention going up and down, and really being able to hone in in the most granular and actionable way.
What are some of the coolest applications of this data you've seen?
I really love the way that music integrates with visuals, and how music contributes to story and tells people how to feel. It's absolutely fascinating to me to look at the way music plays as an extension of that. We've learned a lot about sound; there's a huge difference between someone on camera drinking something with the normal sigh after you swallow and a more exaggerated sound effect that people resonate with.
I also like seeing the difference between how men respond and how women respond, and how we target it differently. What I find really interesting is that sometimes, we think it's going to be men respond one way, women respond to different way, and it turns out it's not men and women. Men and women responded similarly, but some other psychographic breakdowns are different. Something else about their values as opposed to gender, and so being able to see the differences between different audiences and groups are really what we're striving at.
I’d love to hear more about that.
The thing we've learned is that with blanket statements, it’s really hard to say that one group is different than another group because everything is nuanced. So we measured the Wonder Woman trailer and we saw how men and women responded differently, and what's interesting to note is not how they responded differently, but all the ways in which they responded the same.
The other interesting thing was that, when they responded differently, it wasn't necessarily the stereotype of how you might expect them to respond differently. It was something else that would drive those differences. There was an action sequence that men responded really strongly to, women didn't, and so that’s stereotyped as men like action, women don't. But then there was another action sequence that women responded really well to and men didn't. I think the technology allows us to understand that different target audiences will have different responses, but creating blanket statements about each group is just an overgeneralization because the context matter so much.
So, how did you land Michael Eisner and Will Smith as investors?
(laughs) Amazingly, those are two people I never expected to meet, but once I went out and started fundraising, I did the Silicon Valley circuit, and started beating the pavement, there was a lot of interest and it started to happen really fast. One person introduced me to another person, and they introduced me to the other person, and that network effect all unfolded in such a way that I got to meet with some real goliaths in the industry who I admire and never thought I would be meeting—let alone partnering with.
And being supported by Michael Eisner is wonderful; he was CEO of Disney for 20 years, Paramount for 8 years, and he's really a creative genius. He is so intuitive, and he made decisions using his gut instincts, and this was perhaps one of the first times we've seen the research that actually measures the thing that he care about—true emotion as opposed to responses from a survey.
With Will Smith, it turns out he loves data. He's involved in producing major films, and he really understands the value of being able to have a measure that is closer to what a great actor or producer is actually trying to evoke. And that won't be captured with, like, an exit survey at a movie theater.
Where do you go from here?
There are two elements to that. One element is that we need to just hit scale. We need to be able to take what we're doing successfully for every major brand. We need to be the standard brand name who is really owning this new emerging industry of neural analytics, and neural analytics is something that is fledgling so we need to really scale it and own it. We are also expanding globally—we just opened SPARK Neuro Japan in Tokyo.
The other thing is that there are so many applications of our technology, and if you start to think about how this applies to other verticals, that is also critical to our future. We are working on some really interesting work in counterterrorism, which is very important and high-impact.
There are a lot of applications in education that are going to be very important to our future, and ultimately, the true underlying mission of all of this, which I don't really talk about so much because it's not really focus at this moment—but it will be—is the early diagnosis of neurodegenerative disorders like Alzheimer's and Parkinson's and multiple sclerosis. We're collecting more brain data than probably any institution, and that combined with really amazing machine learning capabilities will allow us to understand and advance the science in ways that can only be done when we're able to scale it.
Ultimately, we'd like to leave a really positive mark on the world.