Four Use Cases of Machine Learning in Marketing
With the increasing accessibility of AI, we are on the brink of a world where our inboxes are filled with offers we actually want, our mobile wallets instantly have coupons for nearby stores, and our connected fridge automatically orders more milk. But where does the hype end and today’s reality begin? In this article, Andrea Wildt, CMO of Campaign Monitor, will separate what’s possible now from what’s coming tomorrow, by highlighting the four most effective AI use cases available to marketers today, how they are delivering ROI, and how they are reshaping marketing teams
For decades, marketers have dreamt of personalized experiences at scale. We’ve gotten close — dynamic emails, retargeted advertising, location-based push notifications — but the reality is that configuring and activating these programs on a 1:1 level still takes a lot of manual work.
Machine learning promises to change that, which has us on the cusp of the next great epoch in marketing. This technology will identify exactly how to understand consumer behavior and serve up relevant interactions to drive conversions, greater engagement and ultimately revenue. From entertainment to e-commerce and media to travel, machine learning algorithms are transforming the customer experience. We are on the brink of a world where our inboxes are filled with offers we actually want, our mobile wallets have coupons for nearby stores, and our connected fridge automatically orders more milk.
Aside from changing the entire customer experience, machine learning has significantly advanced (and will continue to shape) the entire field of marketing. In the past 20+ years, the field has shifted from a story-driven approach to a technology-driven one, with marketers expected to bring analytical, data-driven skills to the craft. While many — including Elon Musk — worry of an “AI apocalypse” or a future in which machines will create a risk for unemployment, there are immediate machine learning opportunities for businesses — and, crucially, marketers — right now. Marketers have the chance to redefine their role in setting the technology vision in the company, but we first must separate what is hype from what AI and machine learning is actually meaningful for marketers.
Here are four use case examples:
Predicting user churn to drive greater engagement
While the volume of data captured increases from every new channel or input, the analysis still needs a human touch. With insight into these analytics, marketers can identify how a certain segment of the market typically behaves or predict future patterns. But analyzing this for millions of customers across a dozen (or more) touch points is more than any marketing team can handle. Digital growth company Urban Airship, for example, has developed a machine learning algorithm that analyzes mobile customer behavior to help app publishers identify the most loyal users and predict those that are likely to churn. Armed with this insight, marketers can take action across digital channels to deepen customer engagement or invest more in retaining specific customer segments.
Learn More:
What Is Digital Marketing? Definition, Types, Best Practices with Examples
Incorporating bots for improved customer experiences
Marketers are frequently turning to bots, integrated with popular messaging apps such as Facebook Messenger or Kik, to automatically answer questions about package delivery status or other post-purchase requests, reducing the time spent trying to track down answers to FAQs. Post-purchase experience platform Narvar has built a Facebook Messenger bot, for example, that shoe company DSW uses as a shopping assistant. After customers make a purchase, DSW shares personalized shipping information and makes it easy for customers to track their package. Facebook is looking to extend these capabilities even further with its recent launch of Chat Extensions, which will enable the use of Messenger bots in group chats. With bots from Spotify, SnapTravel, and other available integrations at their fingertips, groups can collaborate on a playlist or even book travel together — powerful tools that might soon tip bots into mainstream adoption.
Scaling personalized content
Brand marketers have long relied on stock images, staged photo shoots or extensively produced ads. Today’s customers are skeptical, and increasingly turning to one another for product recommendations. With a social content marketing platform, marketers can tap into the nearly two billion posts shared on social media every day to find real examples of customers using their products. Travel, automotive, and CPG brands are starting to use software to find, categorize, and publish real photos of its customers and products to better personalize their marketing campaigns. Machine learning technology can learn what content performs best — one person images or group images, for example — and prioritize those results.
Demonstrating marketing ROI
In addition to having the skills necessary to put the technology into practice, marketers must also have the ability to understand and communicate the ROI of each new tool used. This is where AI can help. Brands and sports teams are turning to GumGum for its computer vision technology in order to determine the value of their investments in sports. Each logo exposure on TV and social media is captured and analyzed, resulting in a more comprehensive and accurate media valuation of their sponsorships.
This wave of technologies leveraging machine learning is putting more power into the hands of the marketing professionals and enabling a new era of personalization, sophistication and scale. While these evolving technologies are just starting to make their mark on the world, the effect on marketing is undeniable. Over the next few years it will become even more apparent how machine learning can make marketing even more robust, changing the way brands interact with consumers and fulfilling the promise of a more authentic customer experience.
