Artificial Intelligence in Digital Marketing: Segmentation and Personalization
Originally published on LinkedIn
Artificial Intelligence is becoming an integral part of many industries, including digital marketing. Our CRM Director William Flaiz shares his insights on AI’s useful applications for marketers in his four-part series on LinkedIn. Check out part one below, followed by some key Formative takeaways.
E-commerce has changed the retail landscape drastically, and the competition for customers’ attention (and dollars) has never been greater. While online retailers are able to provide the convenience and competitive prices that today’s customer values, the challenge of fostering customer loyalty by providing an element of personalization has been a pain point. By using AI, online retailers can efficiently identify and connect with their target market, creating a more personalized experience than their brick-and-mortar counterparts can offer.
What Is AI?
AI, or artificial intelligence, is the science that lets computers take on jobs that would otherwise require human intelligence. AI has numerous applications for online marketers, including providing a simple way to segment their customer bases and target each customer individually.
Retailers Who Use AI
AI is transforming e-commerce, and businesses of all sizes are using it to maximize their reach, enhance their productivity, and provide an intuitive and personalized experience for their customers. Here are a couple companies that have put AI to work for them:
Amazon offers a prime example of how retailers can take advantage of automated AI capabilities to increase customer engagement and sales, just by using the data that they already have on hand. Amazon’s automated AI system reads digital user reviews and looks at search histories and past purchases, then uses that information to generate automatic product recommendations.
Unlike most ecommerce companies who have specific target markets that are fairly easily defined, online streaming services have customers of every age, ethnicity, income bracket, marital status, educational background, and field of employment. Netflix users rarely give feedback, so how does the streaming service build its lists of recommendations? Netflix uses a complex algorithm to track not only what their users watch, but also when, where, and on what device they watch it. This data is then compared to that gathered from users with similar viewing patterns. From this data, Netflix provides personalized lists of recommendations.
Three Vendors Who Make It Possible
Founded in 2012 by former Google engineer Amar Chokhawala, Reflektion has developed next-generation personalization software. This software captures site visitor’s’ preferences and behavior and then provides relevant merchandise suggestions and site-search results. On average, this platform raises conversion rates for online retailers by roughly 20 percent.
While online publishing has become increasingly visual over the past few years, the inability to capitalize on a website’s editorial images and video content has been a source of frustration for online marketers. ViSenze developed a solution, using machine learning and computer vision technology to either allow the site visitor to click straight through to the retailer or receive visually similar recommendations.
Founded by Atul Rai, Anurag Saini, Chetan Rexwal, and Pankaj Sharma in 2015, Staqu combines computer vision, deep learning, image processing, and NLP to create a virtual search/hybrid recommendation engine. Basically, rather than experimenting with different keyword combinations and scrolling through endless search results for an article of clothing, a specific refrigerator, or a tile pattern, a shopper can simply upload an image of an item they’d like to purchase, and the platform will perform a search for it and direct the shopper to a site where they can purchase it.
Three Ideas for You: Getting Started
Using AI is an effective way to ensure that your brand is connecting with people who will benefit from it, but knowing where to start can feel like a daunting task. Here are three options that can help you get started:
- Using Existing Software: Many vendors, including Oracle and Microsoft, are beginning to incorporate cognitive capabilities into their software products. If you are already using these vendors, adding AI to your processes may be a smooth transition.
- Building from Scratch with Open Source Software: Google, Facebook, Microsoft, Yahoo, and Amazon have all released open source deep learning and machine learning algorithm libraries. If your company has very specialized needs and is willing to hire a data scientist to use these libraries, then this may be a good approach.
- Picking One Project: Rather than completely revamping the way you operate your business, pick a project that could benefit from AI. For example, Ipsoft’s Amelia, an intuitive, learning, and adaptive virtual assistant, provides an easy way to maximize your call center’s productivity.
A Note on Privacy and Trust
As you incorporate AI into your marketing strategy, it’s important to remember that while your customers value ease and convenience, they also value their privacy. However, most people are willing to share personal information if the payoff is evident and the benefit is clear. As you incorporate AI into more processes, remember to do so with transparency and respect for your customers’ personal information.