Marketing science is an academic field that explores fundamental concepts and specific methods for looking at the market based on objective data and logic. My laboratory actively works with private companies and organizations on joint research to solve marketing issues using actual data provided by companies.

Different companies have different marketing issues. The flow of my research starts with a correct understanding of the issue through discussions with the company’s marketing staff and data scientists, followed by selecting data that can help solve the issue.

We then work to solve the issue by using machine learning and multivariate analysis to analyze the data. Measures are implemented based on research results, and I feel enormous joy when I hear that my research was helpful.

Effective influencer marketing using micro-influencers

Recently, I have been researching ways to use social media and biometric data for marketing. Today, in the context of influencer marketing, there is attention on the existence of micro-influencers with extremely strong influence in specific domains.

In the product domain, I am working to develop a new set of indices that can evaluate micro-influencers from the perspective of companies. How often do they post? Do they have product awareness? Is there a high possibility that the network they have (i.e. their followers) will have interest in a product? These are some of the factors I am looking at.

As for biometric data, I focus particularly on consumers’ line of sight. Using line-of-sight data obtained during the purchasing process, I propose visual merchandising that considers customer characteristics and evaluate relationships between tenants.

I hope that this research will be useful in the future when creating stores in the metaverse that are linked to e-commerce sites. By combining data accumulated on e-commerce sites with biometric data such as line-of-sight data, I seek to realize the provision of products and services personalized for each customer.

Developing data scientists through research activities

Companies conducting joint research with us see merit in the conduct of data-driven decision-making using the latest methods of analysis. When I talk to people from companies, I often hear them say that they are accumulating data but do not know how to use it, or that they are overwhelmed by daily work and cannot verify small ideas within the company.

However, it is within such ideas that you can find elements that are important for proposing marketing measures. Going forward, I hope to collaborate with even more companies.

For students, I believe there is significant meaning in being involved in research that is close to actual work. Among the students that I have guided so far, there were many who sought to become data scientists.

People tend to focus on the glamorous parts, but there are actually many boring areas. For example, before analysis, it is necessary to identify inappropriate data—such as duplicates, deficiencies, and inconsistent representations—and carry out corrections to put the data into a format that fits the analysis.

Such tasks are very dull, but also critical from the point of improving the accuracy of the analysis. I hope people will take time to work with data and grow into full-fledged data scientists through repeated trial and error.

The book I recommend

“Free: The Future of a Radical Price”
by Chris Anderson, Japanese translation by Noriaki Takahashi with supervision and explanation by Hiroto Kobayashi, NHK Publishing

This book explains in an easy-to-understand manner the business models that allow various services to be used for free. I read it while I was a university student as I was very interested in online communications and was deeply inspired by it.

Interviewed: May 2024