AI in Action: Advertising
This is the sixth in a series of articles highlighting the many applications of Artificial Intelligence.
There’s an old saying about advertising usually attributed to John Wanamaker, a 19th century retailer: “Half the money I spend on advertising is wasted; the trouble is I don’t know which half” In many ways, advertising remains a black art — an amalgam of creative and commerce that’s hard to quantify and evaluate. To this day, it’s an imprecise guesstimate as to the value to a brand of running a Super Bowl ad. The rise of new platforms like digital marketing and programmatic advertising have made many forms of advertising more measurable and allowed for an analytics approach to the development and refinement of online campaigns. AI can help in the design of high-ROI campaigns and improve advertising in many other ways, including these:
Promotional Chatbots
To date, most deployed chatbots have come from the customer service department. In the coming years, many other departments are going to start rolling out chatbot solutions to their stakeholders, be them prospects, customers, employees, journalists, or shareholders. There is a chatbot for every public.
An area of incredible potential is applying chatbot technology to advertising and promotional campaigns. Chatbots are a wonderful vessel for strong creative. A chatbot could be used to update or add lustre to an established brand, and also can be used to launch a new product.
A creative team has the ability to give a “personality” or “attitude” to a chatbot that can bring life to a product or service. For example, Pillsbury could have a doughboy chatbot that interacts with their customers and provides recipes. A new Harry Potter movie could have a “Harry” chatbot as well as a “Voldemort” chatbot. Chatbots don’t have to be used to let people check their account balance or find out about new features. They can be entertaining and interactive.
As well as reinvigorating established brands, chatbots can also build new brands. My favorite example of this is Hendrick’s Gin’s Cucumber Adoption Service. Hendrick’s secret ingredient is pickle juice, so the cucumber campaign has a whimsical connection with their product. The FB Messenger chatbot allows you to name your cucumber and “nurture it”. This is a great example of combining good creative with chatbot technology to build a winning campaign. As the creative types get more aware and comfortable with chatbots, we’re going to see some really engaging, interesting applications of chatbots.
Predictive Modeling
The essence of advertising is deciding on how to spend a limited budget most effectively — what mediums, what creative, what vehicles? There are a thousand different parameters to consider. Some examples include:
· Facebook or Google or LinkedIn?
· Should we run a few expensive ads, or a lot of cheap ads?
· What is the best title for our online ad?
· How should we allocate between tv, radio, print and online?
· Should we get involved in sponsorships?
· What is the cheapest way to buy the advertising space we want?
· Who are our buyers? How do we reach them most economically?
There is an infinite number of questions that need to be constantly answered in advertising. Automation can be of great help in some areas such as online advertising — thus the rise of programmatic advertising to automatically optimize your online advertising spend. AI is already being used extensively within programmatic advertising to build models that identify advertising vehicles, and captures non-intuitive approaches to AI. For example, through machine learning its possible to identify markets or users that you would not necessarily consider. A mini sander, for example, might be popular among manicure shops as well as the initial market of woodworkers. These types of insights can be essential in building a thriving business.
Advertising does not lack for data. As well as the traditional sources like Nielsen ratings and Soundscan, there is a huge river of data flowing from the internet every day, including advanced demographics. In the right hands, and with the help of AI, this data can be vital in building your advertising strategy and implementing it effectively. There are no lack of questions, and AI can help you produce better answers.
So in the coming years, AI is going to be used in may interesting ways within the advertising industry. Look out for engaging chatbots from your favorite brand, and for the advertising you’re served to be more customized to your particular interests and tastes. Personally, I am really looking forward to the Nigel Tufnel chatbot.
Part I: AI in Action: Healthcare
PartII: AI in Action: Financial Services
Part III: AI in Action: Entertainment and Media
Part IV: AI in Action: Manufacturing and Distribution
Part V: AI in Action: Mining and Energy
Ken Tucker is a business consultant specializing in AI and Analytics.
