Predictive Marketing is a customer focused, data-driven way to approach marketing. The ability to anticipate customer needs has been coveted in retail for quite some time. But what is new in this digital era is the ability to anticipate at the speed and scale of the digital marketing landscape.

Predictive marketing helps brands and retailers evolve from a product only or channel-centric model to a customer-focused model. Now, this is not to say that having a great product or great delivery method is losing importance. Instead, it is integrating customer needs and wants with product strategy and brand differentiators.

Two of the most common and successful ways Predictive Marketing is being used to anticipate customer needs are:

  1. Predicting customer engagement
  2. Using personalized experiences to increase lifetime value

Predicting customer engagement – Ultimately, brands want to convert a browser to a buyer. So, predicting who will buy from a catalog, open an email, click on a display ad, and determine what threshold incentivizes an order, are all factors of increasing customer engagement and are vitally important to the shopper’s journey. Historically, CRM systems and Email systems have measured engagement using RFM (Recency, Frequency, Monetary) modeling or engagement modeling (how much someone has opened or clicked on an email). The problem there is that this is looking backward rather than comparing current customer behavior to future behavior.

While RFM and engagement modeling work the best in retail (high volume of purchases), Predictive Marketing using propensity models (who is the most-likely-to-buy) can greatly improve marketing results. Using RFM modeling does not enable a brand to recognize potential high-value customer before they have made a purchase. Propensity modeling does.

Propensity models compare the shopping behaviors of prospective buyers to the purchase behaviors of current buyers. The closer the behavior of a prospective buyer is to current buyers, the higher most-likely-to-buy score is given. Propensity models can be used to predict most-likely-to-buy repeat customers as well as most-likely-to-buy first-time customers. Predicting likelihood-to-buy for repeat buyers is a lot easier than predicting likelihood-to-buy for first-time buyers because there is a lot more information.

So, many CRM vendors are adding likelihood-to-buy scoring to their capabilities. By identifying the most-likely-to-buy repeat customers, marketers can fashion the right promotion for this ready-to-buy group.

In addition, many Display Ad Networks use look-a-like predictive targeting to help brands acquire first-time buyers. The downside is that acquisition marketing can be very expensive. Selecting a Display Ad Network that has a continuous learning Propensity Model can result in significantly better outcomes and return on ad spend.

Using personalized experiences to increase lifetime value – Predictive marketing can forecast customer preferences and preferred interactions. Predictive models can predict what product recommendations to make to a particular customer in order to win, upsell, or reengage said customer.

Consumers want brands to deliver some type of personalized experience. Whether it is sending an alert about new arrivals that match their interests, a VIP customer recognition, a discount on products that they previously viewed, a birthday message, or a tailored email on products that they recently browsed, customers want to feel acknowledged. Consumers expect brands to respond in a more personalized way and they want to be recognized (and rewarded) for the time that they have spent with brands.

Technology is helping to automate these personalized experiences. The three most common areas that brands and retailers are finding success in automating personalization are in (a) email marketing (automating triggered and behavioral emails, personalized dynamic content insertion), (b) product recommendations (product-to-product, product-to-user), and (c) personalized ecommerce experiences (most relevant products and personalized offers).

Predicting customer engagement and personalizing the customer experience are two powerful ways brands and retailers are responding to the new demands of consumers. The needs of the consumer are changing, the successful brands are the ones adapting to that change.