How artificial intelligence is influencing Unilever’s marketing
Unilever is using artificial intelligence to influence more of its marketing, from processing insights to finding influencers.
The advertiser has 26 data centers across the globe where scientists are using AI to synthesize insights from a range of sources including social listening, CRM and traditional marketing research. Like other advertisers, Unilever hopes the investments fuel a move away from mass reach channels toward more personalized communications that are also cheaper to produce and localize at scale.
Unilever has been using AI and machine learning to sort through structured data within a database for years, but it hasn’t been able to do the same for unstructured data until recently. Unstructured data is qualitative, which makes gleaning insights from content such as text, audio, social media and mobile activity harder. Rather than develop the algorithms needed to do that internally or even partner with one of the larger tech companies, Unilever is working with startups from the U.K., China, the U.S., Israel, Finland and Singapore.
Unilever’s AI ventures analyze information from multiple sources in the content people post and how they react to them. Only information already available in the public domain is used as part of a wider stance that has seen Unilever deploy its pan-European General Data Protection Regulation strategy to markets that have more relaxed data-privacy regulations.
Through AI-powered influencer marketing platform Popular Chips, Unilever has been able to expedite the search process for influencers. The tech helps to detect those influencers with fake followers as well as pairing Unilever up with the right ones based on demographics such as country, age and gender. Unilever found the startup via its accelerator program the Unilever Foundry.
Having that sort of technology has unearthed insights Unilever’s marketers would have otherwise missed. The most extreme example of this is a range cereal-flavored ice-creams under the Ben & Jerry’s brand. It was inspired after Unilever found that there were around 50 songs that featured lyrics on “ice-cream and breakfast.” The insights came at the same time as the advertiser commissioned research into the ice-cream category, which found businesses like Dunkin Donuts were already serving ice-cream for breakfast. One of the AI algorithms Unilever worked with then sifted through those different data sets and revealed an opportunity for creating sweet treats in the morning.
Two years after the Ben & Jerry’s range launched a range of cereal flavors including Fruit Loot and Frozen Flakes in 2017 and other rivals are now doing the same, said Unilever’s head of insights Stan Sthanunathan, who is the main cheerleader for AI within the business.
“AI is helping us to run metaphor analyses as we’re able to look at all these different signals from unstructured data and start to see how the brain processes information,” said Sthanunathan. “We can start to consider those metaphor analyses when we’re working on brand architecture across the company as well how to look at how we can manage our brands better.”
Beyond market insights, AI is also being used to help recruit executives including marketers. The company partnered with AI firm Pymetrics to build an online hub that uses the technology to assess a candidate’s aptitude, logic, reasoning and appetite for risk against the benchmarks for whatever role they have applied for. The second phase of the process revolves around a video interview from the candidate, whose speech and body language are assessed by the AI. It has cut around 70,000 hours from interviewing and analyzing the candidates, per Unilever.
Despite the growing presence of AIs across Unilever’s marketing, it won’t have a direct impact on the agencies it employs. Critics of AI in media have argued that the speed and accuracy advertisers get when they weave the tech into their marketing is offset by the lack of emotional depth and creativity that comes from human experience. It’s a similar stance to Volkswagen, which has previously used AI to plan its media campaigns.
“If agencies allow themselves to be replaced by AI, then it will,” said Sthanunathan. “If you’re able to use AI to create something new from data that has the curiosity and passion that an algorithm can’t replicate, then there will always be a role for agencies.”
More in Marketing
Co-production is a key aspect of Blast’s esports strategy because it means both partners are invested in keeping “Rainbow Six” esports healthy in the long run, even if their key performance indicators for the collaboration might be different.
To accommodate the global needs of the campaign, Quaker created numerous iterations for Canada and Latin America to reflect the way that consumers in those various local markets use the product.
Investors want to profit from life after the cookie.