How to Use AI for Social Listening
Successful marketing aligns with the opinions, desires, and voice of the target audience.
But it can be extremely time-consuming to talk to the number of consumers needed for truly valuable market research. Unfortunately, limited resources stop a lot of companies from conducting the research that could transform their offerings and elate their audiences.
Linkfluence has 250 employees in seven locations across the globe—and they're working together to use AI for better market research with social data.
We talked to Linkfluence Chief Technology Officer Hugo Zanghi to uncover how social listening and AI create the most helpful consumer data possible.
In a single sentence or statement, describe your company.
We deliver business and marketing insights for global brands, though a combination of social media data, artificial intelligence, and human expertise.
How does your company use artificial intelligence in its products?
Linkfluence uses AI to answer key marketing intelligence questions, such as:
“What do people think about my brand?”
“How is my brand performing versus the competition?”
“What are the upcoming trends in my industry?”
“How should I segment and communicate with my audience?”
“What creates affinity and aspiration in each of these segments?”
While traditional methods involve surveys, focus groups, and interviews, Linkfluence uses AI to collect, structure, and analyze data from hundreds of millions of social media posts, resulting in less bias, more statistical reliability, and unprecedented speed.
Linkfluence uses machine learning and NLP (natural language processing) to building sentiment, topic, and emotion models and to process data with these models, as well as computer vision technology for identifying logos, occasions of consumption, emotion, demographic information, and other data points.
What are the primary marketing use cases for your AI-powered solutions?
Linkfluence’s primary use cases include campaign tracking, reputation and crisis monitoring, real-time marketing, influencer identification and auditing, brand equity analysis, competitive benchmarking, customer segmentation, and trend forecasting.
What makes your AI-powered solution smarter than traditional approaches and products?
Social intelligence solutions, including Linkfluence’s, easily complement or replace traditional market research for many applications, since the data sources and process result in less bias, larger sample sizes, faster results, and lower cost.
Who are your ideal customers in terms of company size and industries?
Linkfluence provides solutions for helping enterprises transform digitally, not just software. These solutions are delivered not only by the standard software support teams that all SaaS companies provide, but also through customer success engineering, software empowerment, consulting, and marketing insight teams.
Linkfluence is therefore best suited for the complex and more evolved needs of larger enterprises, including brands in Consumer Packaged Goods, Fashion and Beauty, Wine and Spirits, Health and Wellness, Automotive, Banking, Agricultural Science, Technology, and Hospitality and Travel.
What do you see as the limitations of AI as it exists today?
Artificial intelligence is good at answering descriptive questions like, “what happened?” This makes it an excellent tool for everyday marketing needs like crisis monitoring, campaign performance tracking, demographic analysis, and competitor benchmarking.
AI is also powerful at answering questions about what to do or what will happen in the future (prescription and predicative analysis), but its usefulness in this area is a direct factor of the expertise of the human researchers who use it.
For example, AI can be used to process millions of social media posts and find subtle similarities in the text, images, emojis and hashtags of these posts, perhaps finding that 5,000 of these have a subtle similarity indicating an emerging trend. However, researchers are required to interpret what these similarities mean in the context of an industry, as might be the case when predicting which ingredients are likely to be the next wellness trend, or which styles might be the most popular in the upcoming season.
What do you see as the future potential of AI in marketing?
AI has already become a fundamental tool for marketing and sales, one that not only shouldn’t be ignored, but can’t be, because it’s already built into everything marketers do. In the case of marketing operations, AI is already deeply embedded both online and offline, such as in search, programmatic advertising, inventory management, and sales forecasting, for example.
However, because all marketing decisions start with asking a question, the most powerful application for artificial intelligence is marketing research.
Linkfluence has chosen to focus on marketing research in order to help companies evolve beyond slow quantitative and qualitative question-based methods, into a unified method, offering a unbiased, deep view of people that both provides numbers and reveals stories.
Consequently, AI-powered marketing intelligence is becoming the most effective way to both track brand and campaign performance online, but also to find answers to unasked but transformative questions about people and our fast-changing cultures online and offline.
Any other thoughts on AI in marketing, or advice for marketers who are just starting with AI?
Like every technology trend, the AI adoption curve has started with over-promising and over-simplification, and not enough emphasis on the data scientists, researchers, and storytellers required to put this young technology to practical use,
At the same time, AI is the future of marketing, and companies that are slow to adopt it quickly in proven applications will fall behind, either due to competitive inefficiency or a failure to spot new opportunities.
Our advice is to focus on the job to be done, not the tool, and on the areas where AI has demonstrated ability to get that job done faster, cheaper, or better. Adopt early, but within an arrangement that allows for course correction and iteration to adapt to lessons and new applications your organization will encounter along the way.
About Paul Roetzer
Paul Roetzer (@paulroetzer) is founder and CEO of PR 20/20, author of The Marketing Performance Blueprint and The Marketing Agency Blueprint, and creator of The Marketing Artificial Intelligence Institute and Marketing Score. Full bio.