Bryan O'Malley, Head of Technology at Fingerpaint
NOTE: The content below contains the first few paragraphs of the printed article and the titles of the sidebars and boxes, if applicable.
Artificial Intelligence (AI) is the technology that everyone is talking about. The big names in Silicon Valley hold dazzling keynotes to showcase the incredible ways they leverage AI to improve their products, and the mainstream press is littered with stories of AI discovering new drugs, helping to identify tumors, brewing tastier beer, and even driving sports cars on the autobahn.
But what about us? How can artificial intelligence help us improve healthcare marketing?
Fortunately, there are a lot of practical-use cases for AI in advertising. Here are two that healthcare marketers can employ right now to enhance their campaign effectiveness.
It’s been said that modern marketing campaigns produce “an embarrassment of riches” in terms of the data they generate. Putting those riches to use is a common challenge, however. Terabytes of data are difficult to analyze, and human perceptions struggle to see patterns across vast and seemingly disparate data points. Fortunately, this is where machine learning algorithms are making a big impact.
By training advanced machine learning algorithms with data from known outcomes—customer journeys resulting in a copay card download, for example—we can identify the common factors in the outcomes we desire. This insight may then be used to develop models capable of predicting outcomes from future customers.
Models trained on prescribing behavior, for instance, could be used to predict physicians who are likely to move from trialists to full-blown adopters, or active prescribers who are likely to lapse. These probabilities may then influence the cadence, channel, and tone of future marketing communications to reinforce or curb that behavior.
As an added bonus, the training process will also provide insight on which indicators led to the prediction. Does the overall number of campaign touchpoints correlate to higher total prescriptions? Or perhaps the number of sales rep visits is a better indicator? Or the number of web visits? Or video views? Or some combination different things?
AI can help answer these questions, allowing savvy marketers to adjust their campaigns accordingly.
Languages are complicated and often difficult to learn. Even within the same language there can be different dialects and nuances depending on geographical location; slang and contractions that make the conversation feel less formal; and cross-pollination of words and phrases from one language to another. They’re also very flexible, allowing us to convey the same information in a variety of ways.
Most apples are red.
Typically, apples have a red color.
Becoming fluent in a new language is notoriously difficult, yet even the most complex language follows a discrete set of rules—something that computers excel at.
A subset of AI called natural language generation leverages the flexibility of languages to create variations in messaging that adhere to the rules of grammar. For example, it can take an existing email subject line, generate 20 new variations, then test the effectiveness of each by measuring their effect on the open rate.
This same approach can be used to optimize the content in search or social advertising, increase click-through rates on display ads, or adjust the call-to-action on a website. Anywhere that text appears in a campaign, AI has the potential to improve its effectiveness through systematic iteration and testing.
Several companies already provide natural language generation as a service, putting this powerful capability within reach of every marketer.
We may only be scratching the surface of what artificial intelligence can do, but these two capabilities are ready for practical use today. Together, they’ll make your campaign more personal and more effective.
Bryan O’Malley is head of technology at Fingerpaint and may be reached at email@example.com.
Fingerpaint is biopharma’s commercialization partner for analytics-enabled integrated solutions. At its core is an award-winning global team of more than 580 people who are committed to creating and executing meaningful brand experiences for healthcare providers, care partners, and patients. Learn more at www.fingerpaint.com.