In the previous chapter, we discussed customer centricity as the most effective business strategy and marketing approach to achieve the goal of growing customer value. If we carry customer centricity all the way to the individual customer level, we end up at personalization. The aim of personalization is to grow customer value by working to detect, anticipate, and predict specific needs, desires, and preferences, and act proactively to customize our marketing activity to be most effective for individual customers.

This concept is not new. Its roots can be traced back to 1993 when the book The One to One Future: Building Relationships One Customer at a Time by Don Peppers and Martha Rogers laid out many of the one‐to‐one customer marketing concepts that we know today as personalization.¹ That was a few decades ago, and while the view that they presented was engaging, it wasn't really doable at scale in practice yet because the technology had not matured enough. Yet the potential clearly was already there.

Personalization hit the radar again when McKinsey declared it the “holy grail” of marketing in a November 2016 article. Personalization—which McKinsey defined as “the tailoring of messages or offers to individuals based on their actual behavior”—showed compelling promise for increasing marketing effectiveness and was developing rapidly. A growing range of progressive companies was working to enable and adopt personalization.² The required customer data systems, analytics tools, and marketing practices were coming into wide use. Six months later, McKinsey declared that the technology “has finally advanced to the point where marketers can use real‐time data in a way that is both meaningful to customers and profitable for companies.” With this combination of advances in technology and marketing processes, data activation and personalization became nothing less than what McKinsey called the “heartbeat of modern marketing.”³ Two years later, in 2019, McKinsey predicted, “Personalization will be the prime driver of marketing success within five years … Advances in technology, data, and analytics will soon allow marketers to create much more personal and ‘human’ experiences across moments, channels, and buying stages.”⁴ (McKinsey has done a lot of good work on this subject, and so have BCG, Deloitte, Accenture, and many other firms.)

As this timeline shows, what was a breakthrough concept three decades ago has become doable and scalable, thanks to advanced technology for data and analytics, digital media, and marketing methods that allow marketers not only to identify customers and predict their behaviors, needs, and wants but also to reach them individually with relevant messages and offerings. Indeed, over the past five years, personalization has taken off, led now by the use of artificial intelligence (AI) to further advance marketing differentiation in real time across customer touchpoints at the individual level. The era of personalization has arrived, with more and more companies investing in enabling and using it at scale in a growing range of product and service categories. As this chapter will show, while personalization is a mainstay for leading marketers today, there are fascinating opportunities for advancing future practices that are now being developed.

Before continuing this discussion, though, we need to ask: What exactly do we mean by personalization? As I define it:

• Personalization is delivering the right message, offer, or content at the right price to the right person at the right time through the right channel.

That's a lot of “rights”—and it's intentional. It means personalization allows the delivery of the most relevant message, offer, or content to a particular customer—what is most meaningful and desirable for that individual. Relevance is the key that unlocks the potential of personalization. As McKinsey stated in its July 2018 report, personalization at scale could drive 5–15% revenue growth for retail, travel, entertainment, telecom, and financial services companies.⁵ Earlier, Ariker et al., in Harvard Business Review, reported that personalization could increase revenues by 5–15% while reducing customer acquisition costs by as much as 50%, and improving the efficiency of marketing spend by 10–30%.⁶ Given numbers such as these, it's no surprise that chief marketing officers have been very interested in personalization in recent years, and their marketing initiatives increasingly aim to utilize this approach.

Technically, personalized offers can be made through a variety of methods and media—for example, by direct mail. (Ask anyone who has had a baby, turned 65, moved, or listed a house, and they'll tell you of being flooded with direct mail offers.) But today, the primary vehicles for personalization are digital channels, especially email, mobile apps, and targeted content on websites. While personalization involves differentiating customers at the most granular level—Fred is different from Jane is different from Mary—digital channels enable us to scale this approach to personalize messages and offers to millions of customers—individually, simultaneously, and often in real time. Plus, digital channels enable us to do so at a much lower cost. This makes it economical to target and personalize much more specifically than would be feasible through direct mail or other conventional media.

Thanks to data and analytics technology, digital media, and advanced marketing practices, the value of personalization continues to grow. Not only do marketers now employ personalization regularly, but customers have come to expect and like it. According to McKinsey, the vast majority (nearly three‐quarters) of consumers now expect companies to deliver more personalized interactions. In 2021, McKinsey declared personalization to be a key engine driving performance and customer outcomes, observing: “Companies that grow faster drive 40 percent more of their revenue from personalization than their slower‐growing counterparts.” Further, it noted: “Across US industries, shifting to top‐quartile performance in personalization would generate over $1 trillion in value. Players who are leaders in personalization achieve outcomes by tailoring offerings and outreach to the right individual at the right moment with the right experiences.”⁷

The value of personalization is being realized today across multiple industries, which include not only retail, travel, telecom, and financial services but also consumer products, insurance, healthcare systems, pharmaceuticals, medical products and, soon, utilities and even waste management. It's important to note that personalization isn't just for, or being used only by, consumer marketers (i.e., B2C). Businesses marketing to businesses (i.e., B2B) can and do employ personalization, with growing effect; and, again, customers increasingly expect them to do so.

For B2B businesses, personalization often is applied at the account level (think of the account, rather than an individual person, as the customer), the distributor or retailer level (the distributor or retailer as the customer) and, increasingly, through B2B2C personalization. In B2B2C personalization, the marketer personalizes services, messaging, offers, and interactions for the individual end user of the product or service provided to its business customers.

For example, a corporate travel agency whose customers are large businesses with many travelers can differentiate the service, information, and offers that it provides to a corporation's individual travelers. John and Mary are employees of the same company, and both search for flights from Chicago to London and hotels in London, arriving on the same Wednesday and departing on Friday. However, the corporate travel management company knows that John's preferred airline is British Airways and Mary's is United Airlines, that John prefers an aisle seat and Mary likes a window, that John usually stays at Marriott and Mary likes to stay at Intercontinental, that John often calls a travel counselor personally while Mary books through her digital app, and that John never extends his stay while Mary will often extend her Friday stay over the weekend and depart on Sunday night.

Thus, in response to the same booking request, the corporate travel management company can propose a flight, seat, and hotel that meet John's preferences and do the same with a different flight, seat, and hotel for Mary. Plus, it can offer John an incentive to use the mobile app for his next booking (thus reducing costs for both the corporation and the corporate travel management company), but not a hotel offer to stay the weekend. At the same time, it can offer Mary a special offer from the local hotel to extend her stay over the weekend (thus generating additional revenue for the hotel, the corporate travel management company, and possibly the corporation) and perhaps also purchase theater tickets—offers that may appeal to Mary but that John would find irrelevant and possibly annoying. (This assumes that British Airways, United Airlines, Marriott, and Intercontinental are all approved for use in the corporation's travel policy and that the corporation permits personalized marketing and additional offers to its travelers.) This realistic example illustrates that personalization in B2B2C marketing can be a win–win–win, generating greater value for all three parties involved.

B2B businesses are also increasingly enabling personalization by their customers. In other words, a growing focus of B2B marketers is enabling their customers (businesses) to do personalized B2C marketing to their customers. For example, a company that makes gasoline pumps for truck stops (a B2B manufacturer) may build in greater features and capabilities to enable users of its product to direct personalized marketing to their customers (truckers and individual travelers) at the pump.

Tapping the Power of Predictive Analytics

If data and advanced analytics provide the customer insights that make personalization work, the question then becomes: How do companies harvest such insights from their customer data? The answer: predictive analytics.

Predictive analytics are used to anticipate outcomes. As described by Kellogg colleagues Eric Anderson and Florian Zettelmeyer, predictive analytic models enable us to anticipate future business outcomes.⁸ In other words, predictive analytics enable us to determine what is likely to happen. (Predictive analytics aren't the same as causal analytics, which enable us to change the outcome of what happens.)

Predictive analytics are an important tool for marketing today because they enable us to answer, and thus act on, practical questions:

The answers to these questions, and many related questions, help us take the right action (for example, the right product offer with the right message) at the right time to maximize our marketing effectiveness, secure and grow customer value, and produce the intended business outcomes. At the individual customer level, the practical insights provided by predictive analytics enable us to personalize.

Let's take the example of a financial services company that offers a range of products. It offers checking accounts, savings accounts, credit cards, different types of investment accounts and instruments, auto loans, home loans, auto insurance, car insurance, wealth management, estate planning, and more. What product(s) should it offer to a particular customer and when? The more it knows about each customer, their stage of life, their transaction history, what financial products they now use, and so on, the more it can determine the next product to offer and when to offer it. Offering a debit card to a young couple, perhaps bundled with a low‐interest car loan and certain basic investment products, may be very appealing, and yet would be totally irrelevant to that same couple years later when they are affluent empty nesters. At that point, wealth management, tax management, and estate planning services may be very important, which would have not been of interest to them at an earlier point in life. And, along the way, college loans, second home loans, and other offerings would be very relevant at certain points and of no interest at others.

Moreover, at times there were probably indications that the risk of losing the customer (in other words, of customer churn or attrition) was growing. Perhaps the couple began adding credit cards from other issuers offering more compelling features. The couple's charge volume and the company's share of wallet were declining. They were using less and less of the company's services and transacting less often (and doing more with competitors). Proactive marketing activity could be the difference between keeping and losing these customers (and thus between continuing to grow their value and losing their future value). In short, the more the company knows about its customers—their transaction history, behavior patterns, experiences, characteristics, and preferences—and can employ predictive analytics to anticipate and act on specific needs, opportunities, and risks, the more it can secure and grow the customer's value.

What makes personalization so effective? It comes down to one word: relevance. Here's a shorthand formula to capture this:

“Data” here includes predictive analytics and now artificial intelligence. What's important is that while data and analytics enable relevance, it is the relevance—in terms of offer, messaging, and timing—that produces greater effectiveness and value.

Here's a simple example. Imagine that the marketing team for an Arizona golf resort is marketing to me personally. My age, income, stage of life, travel patterns, and other characteristics fit the profile of a golfer and suggest that I am an ideal target for the Arizona golf resort. Despite all that, I am not a golfer. I don't enjoy golf. I definitely would not travel to a high‐end resort to play golf. Thus, despite the fact that based on customer segment and profile I seem like an ideal prospect, if the golf resort targeted me with a great offer for a golf stay in March, it would be totally irrelevant to me. They can offer me a discount, a suite, a free massage, free breakfast … I don't care and am just going to ignore it.

Now suppose this golf resort knows more about me individually, including that I love to take my young grandchildren to vacation destinations. Let's say that based on this insight they change their targeted messaging for me to say: “Our resort isn't just a great place to golf—it's also a great place to bring your family. We offer fun and engaging kids’ programs for your grandchildren. Plus, there are lots of other activities and amenities for the rest of your family.” Suddenly that message is far more relevant to me, and I'm interested. I'm likely to pay attention, invest time in getting more information on the resort, check their rates and room availability, and seriously consider a vacation there. And, they may not even need to offer me a discount or other incentives. If they do, the best features to offer would be those that make the stay better for my family, especially my grandchildren.

As noted earlier, not only must the messaging be relevant, but so should the timing. Let's say that someone is an avid golfer, goes to a resort destination with a group of golfing buddies every March, and is a perfect target for our Arizona golf resort. He probably starts planning that trip in about January. Sending him a message in late December, right in the middle of holiday busyness, may greatly diminish the likelihood that he pays attention to it. However, if that same message arrives in the right window—probably in early January—it's going to hit at the right time, with the right message, to produce the intended outcome: that he books a golf trip to the Arizona resort.

This simple example is not at all hypothetical. Thinking about it from the point of view of someone who actually did resort marketing, a series of other questions immediately comes to mind for how to refine the personalized marketing: When did he book last year? How far in advance of his stay did he book? Did he respond to a promotional offer? Did he use other resort services (such as the spa)? Does he search for information on Google? When? What sites does he visit? Does he book directly with us or through an online travel agent? Has he visited our website? How often? What did he search on? Is he a member of our loyalty program? Has he stayed at our other golf properties? And more. Plus, virtually all elements of this example are testable. How far in advance of the intended stay date should we contact the person? What promotional offer will be most effective for whom? What messaging should be used? In actual practice today, marketers are doing exactly this: using data and predictive analytics to personalize and then track, test, and optimize continuously, not just at the customer‐segment level or by profile but for individual customers.

Delta Airlines: Increasing Customer Lifetime Value

To show that actual companies are using these concepts and practices in the real world to grow customer value and thus create business value, let's review the 2018 Delta Airlines’ Investor Day presentation.⁹ One graphic in particular highlighted Delta's goal of “growing loyalty to unlock value”—specifically, customer lifetime value. (As defined in Chapter 5, customer lifetime value, or CLV, is the total of the past and projected future profit generated by a customer.) Delta's SkyMiles loyalty program, particularly in combination with its cobrand credit card, is key to achieving its aim of growing customer lifetime value.

But that's not all. Delta's investor presentation also highlighted its focus on making the right product available to the right customer at the right time with the right offer, a narrative that should sound familiar. In a word—personalization! Delta goes on to specify personalization of offers and service as a key element of its multi‐year strategy to improve all aspects of the customer experience and, thus, grow customer value. Finally, Delta directly relates personalization to increased customer loyalty, leading to greater revenue and profit for the company.

This is a great example of how a leading company in a major industry explicitly described its use of personalization as a means to increase customer loyalty and grow customer lifetime value, which in turn increases the company's revenue, profit, and, thus, its business value. It makes clear that personalization is not just a marketing practice but a key element of the company's business strategy. And, this example is from 2018. The personalization capabilities and expertise of leading firms, such as Delta Airlines, have advanced considerably since then and are still progressing, as we will see.

Auto Mercado: Personalization on a Smaller Scale

You might be thinking that personalization sounds great for a huge international airline with millions of customers, extensive customer data systems, and deep technical resources, but how applicable is it for a smaller company? One of my favorite examples is a regional grocery store chain in San José, Costa Rica, named Auto Mercado. A family‐owned business, it began as a bar and coffee shop established by Guillermo Alonso Rodríguez in 1917 and grew into a regional grocery store chain specializing in imported groceries. Today, there are about 25 Auto Mercado stores located in the San José area and expanding outward in the country. In short, Auto Mercado is a high‐quality grocery store chain, with excellent service and fine products, but not a large company by any means.

Auto Mercado established its successful Auto Frecuente loyalty program in 1998. Around 2017, the company's marketing team wanted to go further and use personalization to strengthen and grow its customer relationships. They built a centralized data platform that captures transactions and other data from multiple customer touchpoints and enables using it to tailor their marketing and customer interactions. By knowing what products the customers in their loyalty program are purchasing, they can send relevant messaging that is individualized to personal product preferences. For example, the company emails recipes to its customers, as many do. In this case, Auto Mercado can send egg recipes to a customer who buys a lot of eggs and let them know when eggs are on sale. On the other hand, a customer who has never purchased any meat or meat products (suggesting that the customer may be a vegetarian or vegan) can receive meatless recipes. It probably wouldn't be relevant to this customer, nor a worthwhile marketing activity, to send them recipes for short ribs and emails telling them that pot roast and hamburger are on sale; it may actually annoy the customer. Again, it's all about relevance. With greater relevancy in the messaging (e.g., recipes and food specials), enabled by customer data on purchase history and other information, Auto Frecuente seeks to enhance the customer experience and increase customer lifetime value through personalization. And, this isn't a global company with a big staff, elaborate customer data systems, and large budgets.

Customer Triggers to Optimize Timing and Content

At this point, let's turn our attention to one particular method of personalization: customer triggers (also known as trigger marketing). As the name implies, triggers are specific occasions or events when a particular message will likely be most relevant to a customer. As such, these triggers are opportunities to optimize the timing, content, offer, and design of customer interactions. McKinsey has found that trigger‐based marketing efforts are three to four times more effective than standard communication.¹⁰

Trigger marketing allows messaging and timing to be personalized based on what's going on in the customer's life—for example, from buying a house to starting a new fitness regimen. But it's not limited to such tangible events. An event can also be a change in behavior, such as purchasing patterns—which can involve buying different product categories or, conversely, decreasing frequency of purchases. Or, the event may have the characteristics of a particular transaction, such as an indication that a consumer is purchasing for a family. The event also may be a series of service experiences with a telecommunication company or electric utility that might involve a certain number of service outages in a period of time. In other words, the event may be revealed in data as opposed to being an observable life occurrence.

I define trigger marketing as:

• Event‐driven marketing practices through which a marketing activity is automatically executed when a specific event or situation occurs.

Some typical examples of events that often are triggers for marketing activity include:

• First order from an online retailer. The retailer may automatically send a welcome message with a promotional offer to incent another purchase soon.
• New mobile app download. The marketer will likely send a message to encourage creating and using an account, perhaps with a promotional offer for referring a friend to become a user.
• A life event. A graduation, a new job, a move to a new place, getting married, having a baby … all of these events are likely to trigger offers from a range of retailers to celebrate, furnish, or get ready for the new development.

You can think of many other examples, and probably have some examples in your email on your phone right now. But, going further, there are other, less obvious examples of trigger marketing events based on high‐propensity characteristics drawn from the customer's profile and transaction history. Perhaps a customer started using a product and then stopped; it might be a subscription service that was canceled or an app that is now seemingly ignored. For example, after listening to a mindfulness app for several days in a row, the person has stopped using it. This often triggers a message such as, “We noticed you haven't listened in a while. Here's some new audio content and exercises that we think you will like.”

An event familiar to most of us is “searched but did not buy”—spending time on a website but deciding, for whatever reason, not to make a purchase. We've all received a message from an online retailer saying, “You left something in your cart.” A more advanced version of the “searched but did not buy” event trigger is when we observe that the same customer has returned to our website or mobile app (or Google) multiple times to search for information on the same subject. If I search for men's athleisure shoes multiple times over a two‐week period, but never buy any, that's a good signal that I may respond to an offer for the product.

Now, let's get into more interesting examples. Decreasing frequency of usage is also an “event” that can trigger a proactive marketing intervention. In this case, it's not that we've lost the customer. It's that the data indicate that we are in the process of losing a customer or that the likelihood of losing the customer completely is growing. (In the meantime, the customer's value is steadily diminishing.) A timely purchase incentive for a product that the customer likes and has purchased in the past, or may like based on previous purchases, can lead to growing customer activity and value instead of losing that customer.

The Timely Reminder

We can use customer data and predictive analytics to reveal changes in a customer's behavior pattern and, thus, anticipate and act on likely customer outcomes and key inflection points in the customer relationship. An example is Audible, the audio book service, which awards subscribers with credits every month that they can use to “purchase” a book. When a customer approaches or reaches the maximum of six unused credits, Audible will send a reminder—a “don't forget to use your credits” message—encouraging people to use their credits for free audiobooks.

A question that came up in a class discussion was why Audible would remind people to spend their credits to get a free audio book that has a cost to Audible. “Why not just let the credits expire and save the cost?” While cost of the digital audio book is minimal, if we imagine that Audible was giving people a free physical book that cost $10, I could still see a rationale for the company encouraging the credit use and choosing to incur the cost. People not using their credits is probably a leading indicator of their not buying any additional books and, possibly, canceling their monthly Audible subscription.

In other words, when subscribers stop taking their free books, it's likely a predictor that they will stop buying paid books, and then not renew their subscription. Diminishing customer engagement and customer value are likely to follow. In other words, it's a signal that the company is in the process of losing this customer, or at least that their CLV is headed downward. The investment in awarding free audio books is well worth the cost to save a customer, stimulate their engagement, and maintain their CLV. Depending on their CLV, a promotional incentive that goes beyond just encouraging a customer to use their earned credits may be a good investment. Would it be worth giving that customer three free credits (perhaps contingent on their purchasing one audio book in the next 30 days)? If the customer is valuable, with lots of future value to come, it may well be. Again, all of this needs to be triggered based on (1) customer data and predictive analytics revealing that not using credits (taking free books) is a predictor of a customer's declining future purchases and/or subscription attrition and (2) the fact that a particular customer has stopped using credits and is signaling disengagement.

Another example is a fitness app such as “MapMyWalk,” which tracks the number and duration of walks taken, distance covered, calories expended, and other metrics. It even provides lifetime stats of total distance, number and duration of workouts, and more. Fitness enthusiasts are probably familiar with the congratulatory messages that pop up after a certain distance or number of workouts. But there may be more to the messaging than seems evident. If a user stops after, say, 13 workouts, that person may get an encouraging “keep going” message. It may seem random that 13 completed walks trigger an encouraging coaching message to keep going. But the company's data may very well show that after 13 workouts (or whatever the number may be), people reach a threshold and their activity starts to drop off—as if telling themselves, “Okay, I'm good; I've done this.” That's the inflection point, predicted from the data, to trigger a personalized message to the customer to get them past that plateau and keep them engaged.

It's easy to extend this example with a lot more specificity. The data, predictive analytics, and messaging can take many more factors into account—the type of workouts; the frequency, duration and intensity of workouts; the profile of the customer; and so on—to personalize the timing and messaging more precisely and relevantly. And, using AI, we could refine this much more.

AI for Personalization

Up to this point, we've talked about the use of data and predictive analytics to inform personalization and identify key events for it. As said earlier, it's all about relevance. Now, let's shift from predictive analytics to AI, and thus to the leading edge of personalization today. AI enables us to scale predictive analytics broadly, economically, accurately, and in real time.

We often hear that AI and machine learning are becoming essential to the practice of marketing and being used increasingly by a growing range of companies. That is true, but one may wonder: What are companies really using AI for in marketing? What are they actually doing with it? The answer is largely personalization. In The AI Marketing Canvas,¹¹ Raj Venkatesan and Jim Lecinski describe how marketers are using and can use AI to optimize their marketing and customer relationships. A theme throughout the book is personalization.

Personalization was among the first AI applications in marketing and is among the biggest. For example, a recent survey of 323 top marketers at for‐profit companies revealed that more than half are already using AI in content personalization and predictive analytics for customer insights.¹² These findings are corroborated by McKinsey's 2021 report about marketers that outperform in personalization. “[These high performers] invest in rapid activation capabilities powered by advanced analytics. Leaders develop at‐scale content creation and AI‐driven decisioning capabilities so they can respond to customer signals in real time.”¹³ AI takes predictive analytics, and thus personalization, to its extreme.

Early adopters have been in this space for several years already. For example, in 2019, JPMorgan Chase said it was using AI to improve the impact and effectiveness of its marketing messages, such as emails to prospective borrowers. The financial giant signed a five‐year deal with Persado, a software company, to put the power of AI behind its marketing copy. In its story about the AI marketing venture, the Wall Street Journal gave a comparison of the promotion offerings composed by human copywriters and the AI‐powered Persado. Humans suggested: “Access cash from the equity in your home” with “take a look” as the call to action. Persado wrote: “It's true—You can unlock cash from the equity in your home” with “click to apply” as the action. Which did consumers respond to better? The data told a clear story. Persado brought in nearly twice as many applications for home equity lines of credit compared to the human‐generated messaging.¹⁴

While using AI and machine learning for personalization is powerful, is it only available to companies with the scale and resources of JPMorgan Chase? What about smaller companies? Is using AI for personalization beyond their reach? No. It's within reach today and steadily becoming more so. AI and machine learning for personalization are available to companies of all sizes from cloud‐based software‐as‐a‐service (SaaS) providers. These include Google, Amazon, and a rapidly growing set of smaller providers such as OfferFit, a firm that offers AI‐based personalization for small and mid‐sized companies. The OfferFit AI personalization engine can learn and optimize the personalization of customer messaging to maximize response and, thus, marketing effectiveness and business outcomes.

Brinks Home Security is an OfferFit customer. According to an OfferFit customer case study, Brinks used the service to personalize contract renewal offers for each customer. Within two weeks, the AI‐generated offers were outperforming the control; within a month, Brinks had improved its profit (measured by incremental customer lifetime value or CLV) by more than 200%.¹⁵ As this compelling example shows, sophisticated AI and machine learning capabilities are increasingly available to and usable by smaller companies, putting powerful personalization tools into the hands of their marketers.

The Connected Strategy

A customer‐centric business strategy (as discussed in Chapter 5) leads us ultimately to personalization. The aim of personalization is to grow customer value through relevance. Data and predictive analytics enable relevance. AI‐driven personalization is now the leading edge of marketing practice. So, what comes next? It is personalization in a world of connected customer relationships.

Nicolaj Siggelkow and Christian Terwiesch articulate the nature and potential of connected customer relationships in their book Connected Strategy. As they describe, we are shifting from episodic transactions to continuous, connected customer interaction.¹⁶ They write: “A connected customer relationship is a relationship between a customer and a firm in which episodic interactions are replaced by frequent, low‐friction and customized interactions enabled by rich data exchange.” Connected customer relationships provide an ongoing flow of information that enables companies to sense customer needs, wants, and opportunities and act before customers ask—often before the customer is even aware of the need. This takes personalization to the next level. The business can personalize its offerings, messaging, and interactions to meet customer needs, solve customer problems, and improve the customer's well‐being before the customer even knows of the problem or opportunity.

The authors outline four levels of connectivity strategies for engaging customers:

1. Respond to desire: As customers search and select products and services, companies respond with reduced friction and increased speed—for example, Amazon's “1‐click” setting to complete a transaction and check out.
2. Curate offerings: Companies recommend products and services with customized and personalized suggestions. An example is Netflix's what‐to‐watch‐next recommendations based on a customer's previous selections.

   The third and fourth levels are where it really gets interesting for personalization going forward.

3. Coach behavior: This includes guidance, reminders, encouragement, suggestions, and even gamification. These can be based on trends and patterns in customers’ behaviors, which the data reveal but customers may not even realize. And, most important, it enables marketers to act prospectively rather than retrospectively. Fitbit is a prime example of continuous customer data enabling coaching behavior. As James Park, founder and CEO of Fitbit, said, “Our users don't want to be told what they did. They want to be told what to do … how to get better.”¹⁷
4. Automatic execution: Now we are at the point of acting before a customer is cognizant, or perhaps even knows, that there is a need. Automatic execution opens a huge range of possibilities in personalization. For example, one can envision continuous blood glucose monitors, as worn today by people with diabetes, which can signal a change, trend, or pattern in blood glucose level that result in a new insulin formulation before the individual is even aware of it: in other words, product personalization and, in this case, personalized medicine. (It's very important to note, particularly for this example, that customer data provision and uses must be subject to and require the customer's knowledge, intentional participation, and approval.)

With connected customer relationships, marketers can move to predictive personalization that will enable coaching guidance and customization of messaging, offerings, products, and services more proactively and precisely than ever before. And these applications aren't limited to business‐to‐consumer enterprises or personal diagnostics.

Consider farming. Deere is a world leader in precision agriculture. Today, a Deere tractor can be a smart technological device that collects a continuous stream of data to “monitor, manage, and maximize” farm operations.¹⁸ In other words, Deere is in a continuous data relationship with the farmer through his tractor, just as Fitbit is with him through the device on his wrist. The day is imminent when changes in soil composition and moisture content are detected as they occur, enabling the customization of inputs from seed to fertilizer, as well as personalized coaching and guidance on farming techniques to maximize crop yield.

The opportunities for advances in predictive personalization will abound in the world of interconnected devices with sensors, known as the Internet of Things (IoT). IoT integration is a natural extension of customer centricity and personalization, and is just in its nascent stages. As one IoT expert observed recently: “In fact, if connected products are aligned with production systems as early as the product creation phase, they subsequently allow the processes themselves to be modified according to actual customer needs.”¹⁹ With personalization extending back into product design and production systems and forward into predictive personalization in connected customer relationships, the growth of customer lifetime value will be enabled and advanced to an unprecedented extent.

The combination of connected customer relationships, providing continuous customer data flow, and AI will advance the practice of personalization to enable a growing range of businesses to accomplish more effectively than ever before the aim of marketing—customer value growth.

Conclusion

Relevance is the key to personalization, which is now doable and scalable for a broad range of companies, using advanced technology for data and analytics, digital media, and marketing methods that enable identifying customers, predicting their behaviors, needs, and wants, and then reaching them individually and economically with relevant messages and offerings. Predictive analytics can personalize, track, test, and optimize continuously at the individual customer level. Trigger marketing, automatically executed when a specific event or situation occurs, effectively targets individuals when a particular message is likely to be most relevant, affording opportunities to optimize the timing, content, offer, and design of customer interactions.

Customer data and predictive analytics reveal changes in a specific customer's behavior pattern or situation and, thus, anticipate and act on likely customer outcomes and key inflection points in the customer relationship. Artificial intelligence (AI) enables us to scale predictive analytics broadly, economically, accurately, and in real time. Connected customer relationships facilitate personalization by providing an ongoing flow of information that enables companies to sense customer needs, wants, and opportunities—often before the customer is even aware of the need. The combination of digital channels, AI, and connected customer relationships is advancing the practice of personalization to enable businesses to grow customer value more effectively than ever before.

Author Biography

Thomas F. (Tom) O'Toole is the associate dean for Executive Education and clinical professor of marketing at the Kellogg School of Management at Northwestern University. He previously served as the executive director of the Program for Data Analytics at Kellogg. His work and teaching focus on customer value growth and related subjects. He developed and teaches a popular Kellogg MBA course on customer loyalty strategy and practices. He is the author of “Branding Services in the Digital Era” in Kellogg on Branding in a Hyper‐Connected World (Wiley, 2019). O'Toole is a senior advisor for McKinsey and Company. He has served and currently serves on the board of directors of public and private companies in a range of industries. He writes for Forbes on subjects spanning academia and business. Until his retirement, O'Toole was chief marketing officer of United Airlines and president of its MileagePlus business unit. Before United, O'Toole was chief marketing officer and chief information officer of Hyatt Hotels Corporation.