The Most Confident Fan: A Glimpse into the Future of Personalized Marketing


In a recent social media campaign, Havas Cognitive analyzed a massive amount of Facebook and Twitter posts with IBM’s supercomputer to determine the confidence level of football fans. The technology could be used in other areas of marketing as well, and help to create campaigns that are better tailored to individual consumers.

IBM has been working together with the global advertising agency Havas Group for over twenty years. A few months ago, the two companies launched a new division called Havas Cognitive. Its goal is to help marketers leverage data more effectively from social media posts, consumer reviews and other unstructured sources.

The new division applies the cognitive computing power of IBM’s supercomputer, Watson, to gain insights that its partners can use to create campaigns that are better tailored to individual consumers. The ecosystem director of IBM Watson believes that the technology will open new possibilities for marketers trying to understand and engage customers.  

“It’s a system that understands natural language, it understands the nuances of what we say, why noses run and feet smell… It understands idiosyncrasies of how we speak, and it also reasons to sort through massive amounts of information and come to hypothesis based on the information,” Jodie Sasse told Fast Company.

The most confident fan


Havas Cognitive is using the system’s beta version to develop programs for brands like Adidas and Red Bull. The very first company that launched a campaign that took advantage of Watson’s computing power was TD Ameritrade.

The American online stockbroker intended to stand out among the sponsors of the NFL and increase social media engagement with its new campaign. With the help of Havas Cognitive, it developed a unique competition for football fans to promote its message that the key to being a successful investor is confidence, and to prove that success and confidence are linked.

In order to win tickets to the Super Bowl, the participants of the contest had to log in to the campaign’s web application with their Facebook or Twitter accounts and show support for their favorite teams in social media.

The winners were not drawn; those football fans who would win and travel to the championship game were the most confident during the competition. Havas Cognitive decided who they were by analyzing  their Facebook and Twitter posts based on criteria such as ambitiousness, self-assuredness, sociability, competitiveness, determination, and positivity.    

The division assumed that teams with more confident fans have a higher chance to win, and they were right. The algorithm correctly predicted that the Denver Broncos would win the Super Bowl, while the majority of sports analysts anticipated the Carolina Panthers’ victory.

Besides proving a point, the campaign was also able to reach its original goals. It won several awards, increased brand awareness, and engaged football fans, many of whom spent a long time on the campaign’s site.   

A different version for every visitor


The technique that Havas Cognitive applied to determine confidence levels, the psycholinguistic analysis of writing, could be used to achieve many other marketing objectives as well.

The division is running tests now to figure out how to alter the tone of marketing messages and change the individual copy of digital media based on what kind of people are reading it. The ultimate goal is one-to-one segmentation on an individual level, even though the company is admittedly not there yet.

While in the future supercomputers might be capable of crafting personal messages to each one of your prospective clients, you don’t have to wait for that to happen if you’d like to personalize your brands’ communication. And you don’t even need Watson’s computing power to move into that direction.

There are many tools available that can help you understand your audience on a deeper level and create content that reflects the personal and professional preferences of your clientele.

Actually, we are developing one of those tools and were able to help many of our subscribers increase the number of conversions by displaying different versions of their websites based on the social profile of their visitors.

The success of campaigns like “the most confident fan” indicate that the web is moving to a more personalized form, and social analytics is going to have a major role in future. If your site offers a wide variety of products, or targets different groups of consumers, analyzing your visitors’ characteristics and experimenting with personalization tactics can pay off big time.

#sharingiscaring 😉

by Péter Szántó
on Tuesday, November 29, 2016

Be the first who reads our next article!

Weekly selection of the best personalization, business intelligence and conversion optimization articles.

It's time to check your inbox! ;)

In order to continue, you have to accept our legal conditions.

If you want to sign up without facebook, you have to provide an email address.

With subscribing on our weekly newsletter you are agreeing to the Terms of Use and Privacy Policy.