Targeted Marketing: Predictive Analytics and Adaptive Content
June 17, 2014
“Imagine the following scenario,” writes Mars Cyrillo. “You are an airline company and from all the data collected from previous interactions as well as data from other sources like social media channels, you are able to identify that ‘Mr. Draper’, visiting your site right now, is not only searching for a flight to LA, but he’s also definitely going to buy his ticket. However, he always searches for prices on your website and ends up buying his ticket somewhere else. What if you could detect that and adapt the search results so that you gave him an incentive to close the deal right there? It could be a special treatment during check-in, a cheaper upgrade, a slightly higher discount, or even a small in-flight convenience – something he won’t find at kayak.com. The more you know about him, the higher the chances that it will appeal to him, irresistibly.” [“Predictive Analytics and Adaptive Content: The New Digital Marketing Transformation,” in, 26 May 2014] In the situation described by Cyrillo, analytics creates a win-win situation for the company and the client. That is the ultimate objective of targeted marketing. What’s unique about the above situation is that the content offered to the client adapts to make the sale. Cyrillo continues:
“Very few people have put the technology and expertise together, but everyone knows it can change the marketing landscape just like computers did. … To an extent, some companies are doing similar things by offering you choices that make you believe they have learned something about you. However, most of it is still rule-based. So while the company might be positively surprising you, they are probably disappointing many other people being treated as if they were like you. Humans are just too complex to fit basic rules.”
As an example of how close the marketing industry is coming to providing customers real-time content, read my post entitled “Real-time Bidding and Targeted Marketing.” Real-time Bidding allows marketers to bid for the right to present an ad in the short time it takes a user to load a web page. That’s pretty fast. The content of those ads, however, has already been determined. Cyrillo is talking about the ability to adapt content within the decision cycle of the user. Cyrillo writes, “Despite still being rare, there are currently cases of software running in the cloud that collect personalized – not personal – data and then find patterns that will reveal a user’s most probable intent from a complete spectrum of possibilities.” Cyrillo argues that programs using predictive activities and adaptive content are the best way to close the gap between customers with high expectations and companies desiring to secure a sale. He believes that such solutions will be the natural next-step in the marketing sector.
According to Cyrillo, three of the four necessary pieces to create a next-generation system are already in place: high-expectation consumers, devices that generate data, and eager retailers. The missing piece involves cloud services and machine learning systems that ingest relevant signals, create models about consumers, and offer up adaptive content based on those models. Analysts at GetSmarter, an online education company, note that obtaining the first piece of the puzzle – data – isn’t a problem. “Smartphones, tablets, PDAs – almost everyone, everywhere, has access to online content,” they write. “Besides by-the-minute status updates, this access also means data – lots of it.” [“Data: the content marketing silver bullet,” Bizcommunity.com, 18 May 2014] They go on to note that this data is “a potential goldmine for content marketers who can translate the numbers into insights about their audience.” They then ask, “So why should content marketers use data?” Hopefully, Cyrillo’s scenario provided the answer to that question. The GetSmarter analysts offer three more reasons: improving market segmentation; making mass customization a reality; and getting insights into consumers’ future behavior. Concerning the first topic – improving market segmentation – they write:
“The data may not be lacking, but the shortage of effective data analysis makes it difficult to make better decisions about the products and services that companies offer. Modern marketers can slice and dice consumer data into meaningful and relevant segments by using Big Data to create useable, relevant groups of consumers that they can target specifically. Because the data reveals exactly who you’re targeting, marketing campaigns can be better targeted and therefore be more effective. Not all customers are created equal. Brands can use data to target and reach influencers – those who have clout in their social circles, or the smaller but lucrative markets. Marketers can target sub-groups of their customers – or specific individuals – by combining data from different online sources. For instance, data from social media platforms about consumers’ interests and preferences can be combined with data from a mailing list. This can show the geographical location of consumers and how many potential customers there are within a specific area, or in which regions marketers should target their campaigns.”
Today, this is the state-of-the-art for targeted marketed; the next-step is adaptive content marketing as described by Cyrillo. If you are interested in learning more about influencers, read my post entitled “Do You Know Who Your ‘Influentials’ Are?” Concerning the second reason that content marketers should use data –making mass customization a reality – the analysts write:
“Taking Big Data and breaking it down into sizable chunks helps content marketers make 1:1 marketing a reality. … By studying behavioural patterns, brands can begin to see what type of content is most appealing to consumers. Brands can then use this to inform the types of content they create – the media format and even the mobile platform that they deliver it through.”
To learn more about how marketers are using behavioral data, read my post entitled “Behavioral Data and Targeted Marketing.” The final reason offered by the GetSmarter analysts as to why marketers should use data is getting insight into consumers’ future behavior. They write:
“The best predictor of future behaviour is past behaviour. By collecting data over periods of time, marketers can pick up on popular topics or themes and create consumer driven content. … This means creating content marketing material based on what people are already talking about, rather than trying to win them over with what your brand is talking about. Knowing your consumer is everything in successful marketing. Data analysis tools are opening up new opportunities for content marketers to understand their audience at a deeper level. Following the data trail means a better understanding of consumers, an improvement in the content seen by consumers, and ultimately a better ROI on content marketing activities.”
I agree that data analysis tools are creating new opportunities for content marketers. I also agree with Cyrillo that the next step, adaptive content marketing, will create even more opportunities and even bigger ROIs for marketers. It should make consumers a lot happier since it will make offers even more relevant to their individual preferences and tastes.