Should we all be depositing poop emojis in our drawers at the notion that artificially intelligent robots are coming to take away all of our marketing jobs?
Or should we high-five?
Should we bar the doors and seal the entrance?
Or should we freshen the guest room and welcome the robots in? Should we give them a place to lay their thermoplastic heads and close their soulless eyes?
You’ve heard that they're already into tasks that we thought only humans could do. And that they will take on more jobs that we thought only humans could do.
In email marketing, artificial intelligence (AI) is going to assist us with better-segmented email lists, better product recommendations for customers based on behavior (past and current), more relevant communications, faster A/B testing, better subject lines, and more and more and more magical stuff of wonder and beauty. And, in some cases, it already is doing much of that.
I asked Paul Roetzer, founder and CEO of the Marketing Artificial Intelligence Institute, for a reality check. Paul founded the Institute to help make AI more approachable and actionable for marketers. (The institute's second annual event—Marketing AI Conference (MAICON)—will take place again in July 2020 in Cleveland, Ohio.)
There are two primary benefits of AI in Marketing, Paul says:
1. AI reduces costs by intelligently automating repetitive, data-driven tasks.
2. AI drives revenue by improving your ability to make predictions.
So we've heard that before. But what's it actually mean for our day-to-day lives? Let's discuss.
Ann: We're in an elevator. What's your go-to definition of AI? Your grandparents are there, so don't exclude them from the conversation by getting too technical.
Paul: My favorite definition of artificial intelligence comes from Demis Hassabis, co-founder and CEO of DeepMind. He says it's "the science of making machines smart."
The simplest way to understand AI is to think about it as a set of technologies and algorithms that are designed to make machines smart, to give them humanlike capabilities (vision, hearing, speech, writing, understanding, movement).
Specifically, machine learning—a primary type of AI—makes machines smarter at making predictions.
And when you start to break down what we do as marketers, we are constantly trying to make predictions to influence consumer behavior:
Which list segment will drive the most sales?
What subject line will get the most opens?
What CTA will drive the most clicks?
Which clients will churn?
Which leads will convert?
Every piece of marketing software you use today—ad buying, analytics, automation, content strategy, conversation, email, search, social—can be made better using AI.
This means the software uses data to make recommendations and predictions that continually improve, rather than marketers having to figure everything out on their own.
Ann: Ah. So AI is about augmenting and improving our capabilities, not replacing our tender, creative marketing souls.
Paul: Our creative marketing souls are safe, for now.
The future is marketer + machine. For the most part, AI will augment our knowledge and capabilities while taking over a lot of the mundane, data-driven repetitive tasks most of us creative folk don't enjoy anyway.
Ann: Like what…?
Paul: AI can help us choose topics and write subject lines. It can edit our copy for sentiment, tone, and style as we type. And it can predict the performance of our content before we publish.
But the human writers possess creativity, curiosity, empathy, emotion, intuition, strategy, and, most of all, imagination. Those are things are very difficult to imbue onto a machine.
That said, major media companies, including the Washington Post, the Associated Press, BBC, the New York Times, and the Wall Street Journal are using natural language generation (NLG) technology to write earnings reports, sports stories, and election results, and leveraging machine learning to enhance human writers through the discovery and analysis of data.
And the story is just starting. As we learned with the introduction of GPT-2 in February 2019, machines are getting smarter, and AI is being used to generate content with minimal human involvement.
Ann: So what does that mean for writers working in Marketing?
Paul: I would say writers who take the initiative to understand and apply AI to their craft will be in high demand in the near future.
Ann: How are marketers already using AI… and might not even be aware they are?
Paul: All of us interact with AI dozens—if not hundreds—of times every day in our personal lives through products and services such as Gmail, Netflix, Alexa, Facebook, LinkedIn, Spotify, Uber, and the iPhone.
You don't care that AI powers those experiences, but you do subconsciously appreciate that they make your life incrementally better through the conveniences of voice, prediction, and personalization.
These same technologies are accelerating marketing toward a more intelligently automated future in which smarter (AI-powered) solutions enable marketers to solve problems and achieve goals more efficiently.
Ann: Since we're here at Emma, what's the opportunity specific to the marriage of AI + Email?
Paul: Email is one of the most natural areas for the application of AI.
AI-powered tools exist today that can write high-performing email subject lines, automatically send personalized newsletters, and optimize email send times.
Machine learning excels at making predictions at scale, a talent that creates exceptional value for email marketers trying to guess what messages, content, and delivery method will result in more opens and clicks.
In short, AI can dramatically improve email performance, tackle some email marketing duties better than humans, and free up marketers to focus on bigger-picture email messaging and strategy.
Ann: Fine. But will robots ever write scripts or something like email newsletters in a truly human-sounding, engaging way? Will I ever outsource my own newsletter (AnnHandley.com/newsletter) to RobotAnn?
Paul: Yes. This is already happening.
The technology is still very early, but there is a race to generate human-sounding language at scale. Most applications to date have focused on shorter text, such as email subject lines and ad copy, but organizations like Google and OpenAI are aggressively pursuing capabilities to expand the possibilities of what AI can create.
That said, generating language is hard. While AI will continue to disrupt writing, for the foreseeable future the net effect will be positive for writers and marketers.
Ann: Edge-case scenario: What's the AI-powered technology that blew you away when you saw it?
Paul: One of my favorite examples of a creative use of AI in marketing was the Salvador Dalí deep fake experience at the Dalí Museum in St. Petersburg, Florida.
"Dalí Lives"—made in collaboration with the ad agency Goodby, Silverstein & Partners—created a life-size "Dalí" using a machine learning-powered video editing technique. Visitors could hear stories from his life and take selfies with him.
From the Verge:
Using archival footage from interviews, GS&P pulled over 6,000 frames and used 1,000 hours of machine learning to train the AI algorithm on Dalí's face. His facial expressions were then imposed over an actor with Dalí's body proportions, and quotes from his interviews and letters were synced with a voice actor who could mimic his unique accent, a mix of French, Spanish, and English.
Amazingly, the technology is readily available to create experiences like this. Nathan Shipley, GS&P technical director, said he pulled the deepfake code off GitHub, an online community used by developers to collaborate and share code.
The Dalí installation is a great example of what's possible with AI. But, in order to tap into its potential, you have to understand the technology and what it's capable of doing.
* * *
So back to the original question: Should we freak? No.
Shortly, the machines will take over. But in the best possible way.
Creatives like you and me will not end up jobless, poor, and living in squalor in a fourth-floor walk-up tenement, eating ramen and dented canned goods, stealing our Netflix Wi-Fi from the bodega downstairs.
Instead, machines take over the optimizing, the analyzing, the reporting, the boring, the drudge, and the data. And it means robots become our partners, helping us be sharper, smarter, and better decision-makers.
That means we can all get back to the reason we went into marketing in the first place: To do great and creative things.
To promote products and people and companies we believe in.
To tell stories worth telling.
To nurture and maintain customer relationships.
To celebrate craft and storytelling.
I, for one, can't wait until the robots are (fully) here.
* * *
Want to learn more? The Marketing AI Institute offers the Ultimate Beginner's Guide to AI in Marketing, a comprehensive guide designed to dramatically accelerate AI education through a collection of more than 100 articles, videos, newsletters, courses, and books. Free access here (registration required).