One-to-One Marketing using Big Data and Predictive Analytics
Tom Cruise’s character striding through a mall in the movie Minority Report as he begins his escapade adventure is a gripping scene. But this isn’t about the movie. It is the depiction of digital marketing in the future that I want to draw your attention to. Eye scanners are used to identify the individuals and then subject them to advertisements relevant only to them (they are even addressed by name!) on the screens around the mall. But is the implementation of this vision of the future a possibility today? In this blog post I will explain how it most certainly is.
So, what is One-to–One Marketing?
A consumer’s involvement is the psychological state that motivates him or her to be more aware about a product/brand, an advertisement or a purchase situation. It also indicates a level of personal importance that the person attaches to the acquisition and consumption of these objects. The degree of involvement has a very significant effect on consumer behaviour. The consumer’s behaviour can either be categorized as low involvement or high involvement.
Involvement is directed towards any or all of the elements of the marketing mix. An individual may show involvement towards the product (its features, attributes and benefits), the price, the store or the dealer or even the promotional effort (advertisement, sales promotion etc). Consumer involvement affects the ways in which consumers seek, process, and transmit information, make purchase decisions and make post purchase evaluation.
Considering that firms view higher customer involvement as desirable, there is a growing trend of one-to-one marketing practiced by them. As opposed to traditional mass marketing where customers are bombarded with hundreds of messages about products that are not relevant to them, one-to-one marketing centers on the customer through two forms: customization and personalization. This enables the companies to effectively target the market segment of size one, thereby accomplishing the objective of increasing customer involvement.
When companies decide what the right marketing mix is for an individual, the phenomenon is known as personalization. This kind of one-to-one marketing is enabled by knowledge about the user maintained in a database. This type of marketing is also called nano-target advertising. Facebook ads are a new form that uses this approach, similar to Google AdWords and Yahoo SmartAds which rely on enormous amounts of information that people reveal about themselves on the world wide web.
One-to-one marketing is therefore all about tailoring of one or more aspects of the firm’s marketing mix to the individual customer. It involves communicating with, selling, and servicing individuals by providing a unique and valuable personal experience.
Facilitating Personalized Marketing Using Big Data and Predictive Analytics
The essential intention of marketing is to build a relationship between a customer and a brand. Both the customer and the firm behind the brand derive benefit through this relationship. The customer experiences the brand at a very personal level, even though data about brand perceptions is aggregated for larger populations. Thus there is a trend in moving away from the practice and technology infrastructure geared towards scaling marketing communication.
The steps involved in facilitating personalized marketing communication are:
- the firms selling their products must know the customers that they are talking to
- they must modulate when, where and how they talk based on the knowledge of who they’re talking to
- firms must ensure that their custom message or offer makes it through the right channels to their intended recipient.
Personalized marketing has become a reality thanks to the internet and technological advancements (such as in-memory data management & warehousing, integrated customer relationship management platforms and predictive analytics) allowing demographic and ethnographic profiling of target customers and brand communities. The steps to execute personalized marketing are as follows:
- Capture All Potential Customer Data: The first step for firms to making their marketing campaigns personalized is to collect all the data they can about their customers. Such Big Data would include profile information, browsing or clickstream data, geotracking data, transactions (online or offline), feedback, social media behaviour on tools such as twitter and facebook, responses to marketing campaigns and loyalty programs, etc.
- Perform Customer Segmentation: Segmentation is a way of grouping people or organizations with similar demographic profiles, attitudes, purchasing patterns, buying behaviors or other attributes to help understand customers more thoroughly and thus market to them more effectively. Using the big data sets, one can create more micro-segments of the customers. Dividing customers into segments of 1 is the ultimate level of personalization, and this is now possible using HANA! The following analysis techniques on the collected big data can be used to create individual buyer preferences and behaviour profiles:
Isolate key performance factors linked to long-term customer value.
Use cluster analysis to form homogenous groups.
- Use factor analysis to identify underlying causes and identify patterns of similarity between observed variables.
Perform affinity analysis to identify market baskests, ie.e, items that tend to be purchased together.
- Perform Recency, Frequency and Monetary (RFM) analysis to determine the Life Time Value (LTV) of each customer.
- Create a predictive score for each customer.
- Create individual Marketing Plans: The last step is to create personalized marketing plans for every customer based upon the business objectives and the predictive score. Analysis techniques could be further used to predict customer responses and behaviour resulting from the campaigns. Actual results, responses and feedback should be collected and analysed by comparing with predicted behaviour to further refine the marketing programs.