Is Your Marketing Team Using AI and Machine Learning…Yet?

As artificial intelligence and machine learning get better, they will become essential for content marketing and public relations experts.

When we think of AI and machine learning, we tend to envision a few specific situations. We conjure a picture of the factory floor patrolled by robots that are guided by ever more sophisticated algorithms. We think of precisely targeted marketing campaigns, including political marketing campaigns, which segment the population into the thinnest of slices with the help of Big Data.

But our thoughts don’t readily turn to AI in the context of jobs that seem to cry out for the human touch, a category that includes many of the jobs we’d classify as professional or white-collar. In fact, though, that thinking must change. AI is just getting started, and few, if any, fields will be left untouched as its growth continues.

It all starts with data, the bigger the better, and the notion of “bigger” here means data on a scale that’s orders of magnitude beyond anything that can be managed by mere humans. The daily numbers are staggering. We send 269 billion emails and 23 billion texts every day, worldwide. We tweet 500 million times, post 95 million photos and videos on Instagram and watch 100 million hours of Facebook videos, all in the course of a single day.

This is data on a scale that would overwhelm even an army of humans, but it’s all in a day’s work for machines. For a PR firm tracking coverage of its clients, the explosion of data in the world is both challenge and opportunity: challenge because of its sheer scale; opportunity because of the wealth of information that is potentially available.

Of course, public relations isn’t just a matter of analysis. Its real purpose is to respond to the information at hand or on the horizon, whether by finding ways to communicate a client’s message regarding events and trends or by influencing a given audience with a point of view. This role is a human one. Machines, for all their analytic prowess, aren’t (yet) masters of nuance, and nuance is really the heart of effective PR.

Nuance is also central to content. It’s one of the things, along with imagination, style and fluency, that make content interesting and engaging, and those qualities are ever more critical in a world where your audiences have so many different options for where to focus their attention.

Because of those requirements, AI is not yet ready to deliver decent content creation except in a handful of narrowly defined areas.  Notably, AI has been used to generate company earnings reports and summaries of minor league baseball games — some 3,500 earnings reports per quarter in the United States alone, and 10,000 summaries for minor league baseball games annually.

But it’s not ready to go too far beyond the driest and most numbers-driven subjects, and its weakness shows even when it tackles the kind of task that seems to be squarely in the AI field of play, a task like translating from one language to another.

If you’ve ever run anything through Google Translate, you know that the end result is frequently a minefield of garbled nonsense. You can barely discern what the original version was trying to say. Here, AI falls short despite the fact that the translating engine has a rich template – the original text – from which to work. Despite that template, which should make the task as easy as possible, the software is apt to miss the point.

Often, this happens because the machine-driven translation delivers literal meaning while missing all the idiomatic usage. That’s the very kind of usage that brings language alive, but navigating it presupposes a natural familiarity with language that’s not machine-like at all.

And if humor enters the picture, all bets are off. Machines don’t get the joke. For content to be engaging and entertaining, let alone leavened with a bit of humor, human intervention is still a must.

Perhaps that will change as AI improves, but AI, even if it can’t yet create all the content you need, is nonetheless quite relevant to content marketing today.

Just as the machines are adept at collecting, parsing and disseminating data, they’re inhumanly accomplished judges of the impact of the messages you’re putting out in response to that data. If you want to test your message against real-world reactions and tailor your message to drive optimal results, and both tasks should be high on the agenda, AI can tell you where your efforts are succeeding and where they’re falling flat. It can analyze response according to small, demographically distinct segments of your audience. You, in turn, can tailor your messages at a microscopic granular level.

Still, the human touch is in no danger of disappearing. Responding to the information you receive is yet another job that takes nuance. Tone is critical, and there’s no shortage of PR efforts that, by getting the tone wrong, became stories of avoidable PR disasters. Until the machines get tone and nuance right, nothing will take the place of PR that’s done by an experienced, knowledgeable and perceptive human being.

The robots aren’t ready to take over the industry yet, but they do have a role, and an important one, as tools whose potential has barely been tapped and they will certainly complement human judgment.

Need some human help? Contact us at 650.485.1280.


Contact Us

We're not around right now. But you can send us an email and we'll get back to you, asap.

Not readable? Change text. captcha txt