Recommendation algorithms are best known for their use on e-commerce Web sites, where they generate a list of recommended items based on input about a customer’s interests.
One of the best known examples is Amazon.com, which uses recommendation algorithms to personalize the online store for each customer. The online store radically changed based on customer interests, searches, wish lists, and purchases. It shows programming books to a software engineer, and baby toys to a new mother.
No wonder that when you compare two important measures of Web-based and email advertising effectiveness – click-through and conversion rates – these personalized suggestions perform vastly better compared to untargeted content (such as banner ads and top-seller lists).
Now the framework is so commonplace that even new fall TV shows are being publicized based on “what you like.”
- Like The Big Bang Theory? Try New Girl.
- People like you who watch The Mentalist have also watched Unforgettable.
- You have Modern Family on your DVR, so why not try Man Up!
- You ordered every season of Mad Men on iTunes, so you should watch Pan Am.
We have seen the retail industry more broadly apply recommendation algorithms for targeted marketing, both online and offline. While e-commerce businesses may have the easiest vehicles for personalization, the technology is also compelling to offline marketers for use in postal mailings, coupons, and other forms of customer communication.
In healthcare, one example similar to Amazon’s is the Web site for Edward Hospital & Health Services in Naperville, Illinois. Last year, Edwards started using real-time behavioral targeting to tailor its Web content to current and prospective patients based on individual health needs. It uses consumer and patient data stored in the hospital’s CRM database to interactively and incrementally customize the content presented to individuals to enhance and personalize the consumer “conversation.”
From our pharma marketing viewpoint, I’ve been pondering the health, medical, and wellness applications of such recommendations:
- If you have this condition, you should pay attention to these associated risk factors.
- If you’re taking this prescription, you might consider this companion product/food to make it more tolerable.
- If you are seeing this kind of doctor, you could also benefit from these supportive healthcare services.
These CRM-enabled Web messages could be displayed as dynamically created, real-time content that contains customized copy, imagery and offers for individual visitors.
Most of all, these relevant health messages would create a more personalized experience that could improve patient engagement.

