Digital Marketing and AI

With the exponential growth of the Internet and the use of digital channels (social media, mobile phones, and other connected devices), marketing has evolved into a level of unprecedented complexity. In addition to the standard attributes (name, address, phone number, demographics, product purchases), marketing today incorporates many digital features such as likes, shares, comments, location, device, time, brand engagement, online shopping, and much more. 

Digital marketing campaigns are executed in almost real-time. The volume, velocity, and variety of generated information are beyond the scope and capabilities of humans and standalone machines. So, how does one manage these large swaths of data, make sense of it, and translate findings into actionable tactical activity?  This is where Artificial Intelligence comes into play. 

IBM defines Artificial Intelligence (AI) as the use of "computers  and machines to mimic the problem-solving and decision-making capabilities of the human mind." Put in much broader terms, AI  can be defined as "intelligence exhibited by machines" (Keng L.  Siau, 2017). Today, we rely on AI when using Interactive Voice Systems (IVR), chatbots, and personal assistants like Alexa, Google Assistant, Siri, Cortana, etc.  

Others areas of AI include Machine Learning, Deep Learning, and Neural Networks. The information contained here is intended for novices who are looking to learn more about the use of Artificial Intelligence with Digital Marketing.

AI is about machines learning from existing data and making predictions based on these learnings. In the article Artificial Intelligence in marketing: Systematic review and future research direction, Sanjeev Verma et al. identified several areas where AI could be used. These include:                                  

• Strategy and  Planning (segmenting, targeting, positioning)
• Product Management (matching product to need, design)
• Pricing (respond to demand fluctuations)
• Product Placement (optimize the supply chain, inventory management, logistics)
• Promotions (content placement, social media campaigns)

Let's assume you launched an extensive social media campaign to promote a four-door sedan called A1 from brand G. The goal is to build a social conversation around the car and brand, increase sales and possibly upsell existing owners. Individuals who own a vehicle are encouraged to share a picture of their vehicle along with the hashtag Brand A1 (Ex: #BrandGAI). 

All comments, images, hashtags along other information are aggregated and studied. As part of the AI learning process, key attributes are reviewed, and relevant clusters are created. For example, group 1 may contain blue cars with standard rims with water in the background. Group 2 may have snow, car tires, and dirty vehicle. 

Through AI, individuals identified in group 1 (based on attributes) may receive an ad that promotes brand Gs super-duper chrome rims. For group 2, individuals may receive messaging that focuses on winter care (new car tires, winterizing your car). 

In a nutshell, the information collected was analyzed, and learning took place. Through the use of AI, background environments, car color, type of tires, rimes were all analyzed based on education, and recommendations were made to prospects based on the data they provided. All of this was done by machines and through AI.