• by George Bilbrey , Op-Ed Contributor, October 31, 2016

We all see the early signs of artificial intelligence (AI) and machine-learning-powered solutions in email marketing. A variety of tools can write (or choose) an optimal subject line for you, or determine the correct product or article to feature in order to optimize clicks for a given subscriber. So what does the future hold for email marketing? Will smart machines supplant humans?

According to the Federal Reserve Bank of St. Louis, the kinds of “routine” work that can be assisted by machines (“routine cognitive” and “routine manual”) have seen flat employment growth, while the kinds of employment that are less routine (“non-routine cognitive” and “non-routine manual”) have seen rapid employment growth.

A great study by two researchers from Oxford scored a large number of standard professions on their likelihood of being “computerised” by AI.

Shockingly, “email marketer” isn’t a standard  job category. But the study found that data-analysis-heavy professions like market research  had a greater chance of being computerized, while less analytic, more “creative” jobs had a lower chance of being outsourced to AI.

The study found several skills, relevant to parts of email marketing, that serve as roadblocks to computerization:

Originality: The ability to come up with unusual or clever ideas about a given topic or situation, or develop creative ways to solve a problem.
Fine arts: Knowledge of theory and techniques required to compose, produce, and perform works of art, like dance, visual arts, drama, and sculpture
Social perceptiveness: Being aware of others’ reactions, and understanding why they react as they do.
Persuasion: Persuading others to change their minds and behavior.

So what does this mean for email marketing? I’ll hazard a few guesses:

• Email marketers will have jobs. A lot of what email marketers do require many of the skills that are less amenable to being computerized. The best email marketers develop original creative and programs. They show a high degree of social perceptiveness—they understand the context and needs of their subscribers and craft email programs to meet these needs. Their emails are persuasive.
• AI/machine learning tools will be used to optimize the original, perceptive, and persuasive programs created by humans. We are already starting to see tools to optimize content in individual messages for each subscriber. We will start to see tools that optimize the types of programs that each subscriber sees (e.g., How many messages in an abandoned shopping cart campaign? Should this subscriber get a “happy birthday” message?). We will see more tools that optimize cadence and timing to maximize subscriber lifetime value or revenue.
• This means that email marketers will need to have a deep understanding of subscribers and be more adventurous in providing what subscribers need. Understanding the customer (subscriber) will be key to building persuasive programs. Marketers will need to think more “out of the box” to make a difference.
• Human-driven analytics will be more about understanding whether our high-level program-design choices are working, and less about making optimization decisions. The machines will be doing the optimizing. Email analytics will focus on whether the machines we are using are working (and whether our new programs are paying off).

The future isn’t written, nor is it clear. What is your take?