New research from the University of New Mexico's Anderson School of Management provides data insights into what makes travel influencer content successful on Instagram. Assistant Professor Hyunsang Son has applied advanced machine learning techniques to analyze over 4,000 Instagram videos from the top 40 travel influencers.
The study employs a transfer learning approach—a method used by technology companies like Google and Facebook—to extract and analyze features from videos, including length, sentiment, facial expressions, and visual elements.
"I can actually quantify the exact level of contribution each feature has on the number of likes and comments," Son stated. "It's a combination of computer vision for feature extraction and traditional econometric modeling to estimate effects."
Findings Challenge Creative Assumptions
The research reveals several surprising factors that drive engagement:
- Short-form videos consistently outperform longer content
- Excessive smiling reduces engagement levels
- Emotionally expressive content, including traditionally negative emotions like anger and sadness, increases engagement compared to neutral content
- Product size visibility positively affects user interaction
- Brightness contributes to positive engagement, while excessive contrast or rapid scene changes diminish performance
- Overuse of hashtags correlates with lower engagement
Practical Applications for Marketing Strategy
Son's findings suggest marketers can now provide influencers with data-backed guidelines rather than just creative direction. The research has applications for local and national tourism organizations, which could analyze their own social media content using similar machine learning techniques.
The study is connected to broader marketing concepts, including media richness theory, which suggests that richer media, with expressive emotion and visual detail, generate more engagement than simpler content forms.
"As marketers, we can now offer formulas for influencers: avoid over-smiling, highlight the product size, and use bright visuals," Son noted. "Short videos tend to work better. Too many hashtags? Not helpful."
Son believes this approach marks a significant development in marketing strategy formulation, as it applies machine learning algorithms to better understand human behavior and response patterns.
The full study is available here.
New research from the University of New Mexico's Anderson School of Management provides data insights into what makes travel influencer content successful on Instagram. Assistant Professor Hyunsang Son has applied advanced machine learning techniques to analyze over 4,000 Instagram videos from the top 40 travel influencers.
The study employs a transfer learning approach—a method used by technology companies like Google and Facebook—to extract and analyze features from videos, including length, sentiment, facial expressions, and visual elements.
"I can actually quantify the exact level of contribution each feature has on the number of likes and comments," Son stated. "It's a combination of computer vision for feature extraction and traditional econometric modeling to estimate effects."
Findings Challenge Creative Assumptions
The research reveals several surprising factors that drive engagement:
Practical Applications for Marketing Strategy
Son's findings suggest marketers can now provide influencers with data-backed guidelines rather than just creative direction. The research has applications for local and national tourism organizations, which could analyze their own social media content using similar machine learning techniques.
The study is connected to broader marketing concepts, including media richness theory, which suggests that richer media, with expressive emotion and visual detail, generate more engagement than simpler content forms.
"As marketers, we can now offer formulas for influencers: avoid over-smiling, highlight the product size, and use bright visuals," Son noted. "Short videos tend to work better. Too many hashtags? Not helpful."
Son believes this approach marks a significant development in marketing strategy formulation, as it applies machine learning algorithms to better understand human behavior and response patterns.
The full study is available here.