Choosing is an expensive operation (in terms of time and energies), so people tend to decide quickly what they want to do. Often, if they do not find what they are looking for or do not receive the right incentive, they will abandon their search. Based on this assumption, big companies like Amazon or Netflix have developed product and service recommendation algorithms, with the goal of maintaining the consumer engagement and showing the best products.
Help me deciding
As Jeff Bezos says, “We don’t make money when we sell things. We make money when we help customers make purchase decisions.” And this is so true: the user experience is better if the decision time is also reduced, it is not just a customer journey but also a decision journey.
The users, with their traceable behavior, “release” enough data to expect personalized experiences. Artificial intelligence algorithms therefore are used for this purpose: to propose the most interesting content for the single customer and create a fluid process of choice (avoiding, among other things, that the user focuses his attention elsewhere). It is obvious that this contributes to increase the sales and the customer’s lifetime value; a personalized experience is undoubtedly one of the main components for the customer’s loyalty.
How does an artificial intelligence algorithm work in marketing? The Netflix case.
Following this article (ACM Transactions on Management Information Systems, Vol. 6, No. 4, Article 13, Publication date: December 2015), we can explain how Neflix has used artificial intelligence to improve the user experience. We will describe the idea behind some of their algorithms to understand how artificial intelligence works in marketing automation.
The first question Netflix asked itself was: What information should we consider to offer our clients something they might like? The first analysis, which date back to some years ago, focused only on the rating: they calculated how many stars the users had given to movies and, based on that, the algorithm selected similar products.
The change happened when Netflix started to analyze other elements: for example, when and how movies were viewed (type of device, day of the week, time) how the users found them, etc. They also started to consider which content was recommended, but not clicked, and to use the algorithm’s failure as source of information for the algorithm itself. People leave a lot of information just by using the product… why not using them?
This idea started the creation of several algorithms that “collaborate” with each other’s. The first one is called PVR (Personalized Video Ranker) and it is based primarily on customizing individual user preferences on the entire catalog. The second one is the Top N video ranker that aims to find the best recommendation from the catalog focusing only on the top overall rank of all the users.
All the algorithms, along with the amount of products shown, contribute to what is now called the Netflix Experience.
How does artificial intelligence work in marketing automation?
The machine learning is a continuous learning process. In the marketing environment, specifically, it analyzes all the data the user left every time he interacts with the brand. What do algorithms analyze? The products a user has bought , his purchasing habits (hours, days, time, devices), what he saw on the website, the type of channel he used to contact the company, the emails he clicked and the posts he read on Facebook, as well as many other relevant variables to offer the best customer experience. Today technology allows platforms to collect and standardize all kind of information, online and offline, to improve algorithms every day.
Not all the marketing automation platforms are alike and the big difference lies in the artificial intelligence. Some operate exclusively following pre-established patterns and they work only whether or not a setting is triggered. We can say that they execute orders after the user has done or not a specific action (e.g. e-mails notifications after a click).
On the other hand, those based on machine learning, learn constantly and make decisions instead of people, by aggregating the information in a way that would be impossible for an operator. The artificial intelligence in marketing automation is therefore proactive and uses data not only to react to past stimulations, but above all to predict what the users will want.
Finally, here is a reading about a survey by Smart Insight that explains why the predictive marketing is one of the huge trend of 2017. Did you know that?