AI tools can createbetter look-alikemodels for prospecting,determine the mosteffective contentmarketing, efficientlyset up and interpretA/B email and onlinetests, and help fine-tune messaging.
Artificial intelligence, or AI, used to be the stuff of science fiction. Now it’s science—and marketing—fact. Ifyou’ve ever looked at your website performance via Google Analytics, asked Siri a question, or scrolledthrough a list of movies on Netflix, you’ve interacted with AI. And with the continuing deprecation of third-party cookies, AI is becoming an increasingly important tool for marketers.Once Google phases out its third-party cookies by the end of 2023, a marketer will be virtually unable totrack the behavior of its customers and prospects when they are not on the marketer’s website. Rather than relyheavily on this third-party tracking data to optimize the effectiveness of ad placements, retargeting, and thelike, organizations will need to make the most of the first-party data they own about their audiences. AI canhelp them do just that. By finding data correlations that might otherwise be overlooked—or that could takemuch longer to discover via more manual means—AI tools can create better look-alike models forprospecting, determine the most effective content marketing, efficiently set up and interpret A/B email andonline tests, and help fine-tune messaging. AI can also improve customer satisfaction, along with retention, viachatbots, website personalization, and improved site search.“A refocused and refined first-party data strategy will be paramount in the years to come, and AI, as aprogressive technology, can help make better use of these data assets,” says Julie Saxon, Chief Revenue Officerat IBM Watson Advertising and the Weather Company. “Regardless of how good the data sets are, marketerswill still need technology to help make sense of what’s in front of them. That’s why we believe AI can be thefoundational technology to advancethe data narrative. But now the challenge to both open-web marketers andpublishers is whether we’re willing to take the steps necessary to invest, test, and implement AI widely across ouroperational teams. Are we ready to make AI the industry’s foundational technology, creating deeper connectionswith our audiences, or are we okay with the status quo?
”Are we ready to make AI the industry’s foundational technology, creating deeper connections with our audiences, or are we okay with the status quo?
a similar mindset of recognizing that they are the stewards of their attendees’ time
What Is AI?
Artificial intelligenceconjures up images of machines with minds of their own taking over, à la HAL, theeerily humanlike computer in2001: ASpace Odyssey. In practice, however, it’s not at all scary. In simplestterms, AI applications are those that have been programmed with algorithms to respond to data in a way thatmimics human behavior.Let’s take a form of AI most of us are familiar with: product recommendations. A basic bookrecommendation program would be written so that someone who bought a book by Stephen King would, onsubsequent visits, receive recommendations for other books by King, as well as books on similar subjects asper predetermined parameters. As this example shows, at the heart of AI are if-then and if-else statements.Now let’s say that, over time, a significant percentage of buyers of Stephen King books also bought booksby Danielle Steel. Given how dissimilar those two writers are, the person who wrote the recommendationprogram most likely did not include code along the lines of “if customer buys King, then recommend Steel.”But if the program was written to encompass machine learning (ML), a subset of AI, then it would detectbuying trends such as this and subsequently “learn” to recommend Steel books to King fans and vice versa.A subset of ML—and therefore a sub-subset of AI—is deep learning, which requires large data sets andsophisticated programming to solve for complex problems. Self-driving cars, Alexa, Siri, and Netflixrecommendations are among the applications dependent on deep learning.Natural language processing, orNLP, is another subset of machine learning. This enables computers to interpret and analyze human language,not just code. Chatbots and voice search applications are among the more common instances of NLP in action.
AI in Use
The alphabet soup of acronyms can obscure the wealth of ways artificial intelligence can help organizationsmarket more effectively.
Below are examples of five common AI applications:
Product recommendations. According to a study by e-commerce solutions provider Kibo, 52 percent of retailers already have a product- recommendations engine on their site—and for good reason. Suggesting relevant products to site visitors can not only convert visitors to buyers, but also increase average order values. Businesses have also successfully used recommendation engines to personalize abandoned-cart and post-purchase emails, boosting conversions and leading to add-on sales.
Build with Ferguson, which sells home improvement products to both consumers and trade professionals, realized that it needed to address these different audiences in different ways. Working with optimization solutions provider Dynamic Yield and its AI-powered segmentation tool and recommendation engine, the company tested multiple suggestion algorithms. It found that recommendations based on recently viewed products were most productive among trade professionals, whereas recommendations that also incorporated affinity algorithms, based on purchases made by other shoppers with similar interests, generated the best response among consumer shoppers. By better personalizing its suggestions, Build with Ferguson saw an 89 percent jump in sales of recommended products. What’s more, it found that site visitors who interacted with product recommendations spent 13 percent more than those who didn’t.
Marketing messaging. Marketers have used A/B testing for decades as a way of fine-tuning everything from the most effective time of day to send emails to the best color for a CTA button. Savvy marketers are now using AI to expand their multivariate tests and generate actionable results more efficiently than they could before.
The nonprofit charity: water, which is dedicated to giving people worldwide access to clean water, turned to Persado’s AI platform to boost the effectiveness of its Facebook ads. The technology generated 1,024 permutations of 16 ads—far more than could have been created and tracked using more manual means. By testing these variations, charity: water was able to determine which combinations of imagery and messaging were most productive among a variety of audience segments. The most effective ad proved to have a 227.4 percent higher conversion rate than the least effective. Implementing the most productive ads resulted in a 20.6 percent lift in views of charity: water’s website content and a 32 percent jump in donors across all audiences. It also provided the organization with insights it could use in its other marketing efforts: for instance, messaging focused on community (“Together, we’re unstoppable”) was more effective than imperatives (“Click here to see how you can help us make history”).
Email segmentation. The death of email as a marketing tool has been predicted for years. Yet email marketing platform provider Litmus estimates that the medium generates $36 in revenue for every $1 spent, with ROI increasing to 45:1 among retailers, e-tailers, and CPG brands. If your email marketing efforts don’t generate similar returns, AI can help in several ways. Programs can segment your email list in ways you might not have considered so that you can better personalize the messages individuals receive, rather than sending one-size-fits-none blast emails. AI tools can also facilitate multivariate testing, similar to what charity: water did with its Facebook ads. And content-focused AI applications can draw on algorithms to help you hone your subject lines, offers, and overall language.
State & Liberty, a multichannel seller of apparel for men with athletic builds, had been manually creating customer segments in hopes of determining the optimal send times, frequencies, and content to send to each. Once it implemented Retention Science’s AI-powered Cortex email platform, State & Liberty was able to quickly create well-defined audience segments based on where individuals were in the customer life cycle, among other traits, and could appropriately vary the emails sent. Not only did the AI tool save the marketing team time, enabling them to focus on other projects, but it also generated a 73 percent lift in revenue, a 26 percent rise in conversion rates, and a 25 percent increase in average order value.
Chatbots. Nearly one in four consumers used a chatbot to communicate with businesses in 2020, according to conversational marketing solutions provider Drift—almost double the 13 percent that did so the previous year. Chatbots are software applications that rely on AI to conduct online or text-to-speech conversations with individuals, and they are most commonly used as virtual customer service agents. They can answer simple requests on their own as well as gather relevant information from a consumer to pass on to a human agent for resolution of more complex issues. Successful chatbot programs help organizations cut costs—by up to 30 percent, according to IBM—by reducing the number of human contact center agents needed per shift, even providing coverage during hours when there are no live agents. And because they enable consumers to “self-serve” regarding questions and issues, they reduce the number of call center requests.
MSU Federal Credit Union implemented chatbots to improve customer service, but not in the way you might expect. Rather than have AI-powered agents interact with customers, the credit union worked with solutions provider boost.ai to program a chatbot to interact with its own live agents. In just 10 days the bot was programmed to answer questions about the credit union’s most popular topics. In the past, when customers asked questions that the live agents were unable to quickly answer, the agents put the callers on hold while searching their knowledge base or asking a colleague or a manager. Now the live agents could ask the chatbot and quickly get the answer. What’s more, machine learning enabled the chatbot to become even more effective over time. After one month, the credit union and boost.ai estimated that the chatbot was able to eliminate 2,000 employee-to-employee interactions, dramatically improving contact center efficiency.
Competitive analysis. The more competitive a market segment, the greater the need for competitive analysis—and due to the sheer number of competitors, the more labor-intensive conducting such research and analysis is. Upserve, which provides point-of-sale and management software to restaurants, implemented Crayon’s competitive intelligence solution to help it stay on top of the ever-changing market. The software uses AI and ML alongside human learning not only to identify its competitors’ latest products, services, offers, and pricing, but also to guide how the company could best react to win a greater share of the market. Upserve used the tool to help determine which changes it should make to its product offerings as well as how and when to update its sales and marketing materials to show its offerings in the best light compared with its competitors. As a result, its client win rate against its top five competitors rose 54 percent. The solution also saved Upserve’s senior product marketing manager eight hours of work a week.
Not Just for the Big Guys
As these examples show, you needn’t be a billion-dollar business to have access to AI. Yet according to ITindustry analyst firm Techaisle, only 22 percent of small businesses in the United States (those with fewer than100 employees) plan to adopt some form of AI, and just 5 percent are already doing so. Among midsizebusinesses (those with 100–999 employees), 28 percent plan to implement AI, while 26 percent already use it tosome degree.One factor preventing SMBs from adopting AI tools is cost—or the perceived cost. However, the rapidincrease of providers and solutions has resulted in a plethora of options available at a vast array of pricepoints. A more significant reason might be fear of the unknown and misconceptions of howaccessible andpractical AI really is. “For many, the termAIis a black box that they don’t understand,” says Tink Taylor,Founder and President of marketing solutions provider dotDigital. “People like what they know, and theydon’t know the new marketingplatform products like recommendations or predictive analytics.”Yet, Taylor says, “AI is becoming—and will become—more and more central to marketing platforms.”So, ignorance is not bliss when it comes to artificial intelligence. Organizations thatfail to integrate AI intotheir marketing efforts, especially as more of their competitors do, might find ignorance more costly thanimplementation