To meet the growing demand for each customer experience to be tailored to individual preferences and tastes, marketers are quickly realizing they need to facilitate personalization across all marketing channels.

As our society becomes increasingly Web-engaged, users are growing accustomed to the personalized experience that Web 2.0 provides. With constant online access through smartphones and tablets, customer demand is starting to reflect the personalization perpetuated by the filtering, following, and friending options available on email, Twitter, Facebook, and LinkedIn.

Those capabilities allow users to apply their personal preferences to control what they want to hear and from whom. Those techniques have moreover bred the expectation that the same level of personalization should be available everywhere.

Delivering only relevant messaging to each customer is one way marketers can satisfy customers' growing demands and avoid wasting resources. By not reaching out to uninterested customers who will likely ignore unwanted information, or worse, become annoyed by it, marketers can save money and increase efficiency.

A sophisticated analytical technique called uplift modeling can empower organizations to target only those customers who will be positively influenced by a marketing campaign, helping to reduce campaign costs, improve customer experience, increase customer lifetime value, and reduce churn. More on uplift modeling later; first, a few words about why email is anything but free.

The Hidden Cost of Email

Though mail merges and email blasts seem a quick and "free" way of addressing a company's entire customer base by name with information about a promotion, new product, or service, they are becoming less effective—and, in some cases, detrimental to the success of marketing campaigns.

By sending untargeted generic emails intended to reach as many customers as possible, marketers lose the opportunity to engage customers on the personal level that those customers now expect. In addition, marketers risk driving customers away by contacting them about irrelevant products or services.

Companies risk enough when they send just one untargeted email to customers, who might merely dismiss it in the first instance, but sending further irrelevant emails could leave customers frustrated and damage a positive relationship—and potentially reduce customer lifetime value. Every untargeted message they receive serves as another reminder that the company does not understand them.

Even if frustrated customers don't physically opt out, if they're receiving frequent emails from a company about products or services they've already purchased or have no need for, chances are they'll mentally opt-out when they see an incoming message from that company. For those customers who don't manually opt out the first time, every irrelevant email they receive is another invitation to do so and take their business to a competitor.

Once a customer chooses to unsubscribe, the company loses both the ability to contact him or her, and the opportunity to derive future sales revenue.

Identifying the Right Customers to Contact

How can marketers ensure that they're conducting effective email campaigns and targeting the right customers with the right messages?

Uplift modeling, also known as incremental or lift modeling, is a successful segmenting technique. It predicts the difference or lift that a marketing campaign will have over what would have happened in the absence of the campaign.

Today, marketers use business intelligence and data mining tools to cut and slice historical data to reveal patterns that will inform marketing decisions. However, that approach tells them only what has happened. Future interpretation is left up to the skill of the analyst.

At the next level, predictive analytics provides a window into the future. Given an objective, such as propensity to churn or respond to a product offer, predictive analytics will generate a behavior model that will help organizations predict future customer behavior. From that, they will know who is likely to respond to a campaign—but not who has been positively affected by the campaign.

Uplift modeling, however, clarifies exactly that distinction, empowering companies to segment and target only those customers for whom marketing can make a positive difference.

By simultaneously modeling customer behavior from treated and untreated segments (control groups), uplift modeling is able to segment potential campaign respondents as follows:

• Persuadables: People who buy (or renew) and who would not have done so had they not been campaigned to
• Sure Things: People who would have bought whether or not the campaign ran
• Lost Causes: People who would never have bought, with or without the campaign
• Sleeping Dogs: People for whom the campaign triggers a negative response (because of being annoyed, being reminded of alternatives, or simply being given a chance to overcome inertia)

This sort of segmentation allows marketers to focus efforts only on the Persuadables, those who are likely to respond positively to a marketing campaign. Similarly, uplift modeling prevents businesses from wasting time and money targeting customers who are already a Sure Thing and who will buy regardless, as well as those who are a Lost Cause and will never buy. Uplift also helps marketers avoid the Sleeping Dogs, those who don't want to be disturbed and are likely to react negatively to marketing outreach by opting out of marketing lists, or even taking their business elsewhere.

Using this next level of customer-response segmentation not only reduces the amount of money that organizations spend on outreach but also helps reduce a company's cost of fulfillment, as it prevents marketers from giving away special offers or incentives to customers who were going to buy or renew anyway—at full price.

The insights provided by uplift modeling dramatically increase a campaign's return on investment by allowing marketing teams to target fewer people yet still produce a higher response rate and increased sales.

As companies have been forced to adapt their marketing outreach to customers' growing demand for personalization, highly targeted campaigns have become a standard necessity. If they haven't already, companies should look at replacing their mass, blanketed outreach with messages tailored to each customer and based on specific customer data.

By incorporating intelligent analytics into their marketing strategies, organizations can ensure that their messaging is customized to each consumer, thereby achieving better campaign results, while building customers' trust and improving their loyalty to the brand.