AI Will Help Digital Marketers Work Smarter, Not Harder. Here's How.


AI is not the future. It’s already here.

It’s not quite Skynet, but we’ve made some pretty remarkable advances in artificial intelligence since IBM Deep Blue upset world chess champion Garry Kasparov in 1996. It can seem painfully esoteric, as evidenced a couple of weeks ago in HBO’s satirical comedy Silicon Valley.

Before getting into jargon about neural nets and natural language processing, let’s define artificial intelligence. In digital marketing, usually we’re talking about a specific type of AI called machine learning. This involves a series of APIs that allow a program to take in unstructured data (basically, anything that the program has never seen before or has never been specifically programmed to understand) and “learn” how to process that data in ways that mimic human intelligence. This can include visual or speech recognition, language translation, even decision-making such as risk analysis.

Often, machine learning involves natural language processing. This allows software to understand conversational, plain language and even respond in ways that would make you think you’re talking to a real human. It doesn’t always work, but you get the idea. Natural language processing is appealing to marketers because of the ways it allows them to interact with customers, clients, and consumers.

Knowingly or not, odds are you’ve interacted with a chatbot at some point. Maybe you were returning a purchase or disputing a billing error. Maybe you reached out for more information about a service or product. Or maybe you were in China on business and tried to book a flight through WeChat. Some interactions are hilariously cringeworthy. Other times you don’t even know you’re talking to a computer. Just recently, a professor at Georgia Tech used IBM Watson as an online TA and students had no clue.

Chatbots and conversation-based UI will revolutionize how others interact with your brand.

This technology is everywhere. For Mothers Day, 1-800-Flowers unveiled an AI concierge service that would allow customers to find the right gift for their moms using ordinary conversational language. There is a Slack bot integration that will order your lunch. Facebook has even begun rolling out artificially intelligent chatbots to Facebook Messenger, where they can answer basic questions about products, services, or events on behalf of Facebook brand pages.

Products like Amazon Echo and Google Home are already taking things one step further, from a text-based chat environment to one that is entirely conversational. If you find yourself out of coffee filters, all you have to do is say so within earshot of one of these devices and it will instantly re-order it for you.

It’s kind of creepy, but also kind of cool. That is, of course, until one day your Echo revolts and responds to your request for more toilet paper with “I’m sorry, Dave. I’m afraid I can’t do that.”

We’re still in the early days of machine learning, but we’re already seeing more streamlined processes and workflows, better use of real-time data, and more efficient ways to interact with and serve customers.”

Image recognition technology will make ecommerce even more convenient and efficient.

Imagine an app for a B2B vendor that allows customers to automatically reorder office supplies, business products, or merchandise simply by taking a photo of it when inventory is low. Similarly, Flow, an Amazon-powered app, allows consumers to snap photos or scan barcodes of any product and find it on Amazon along with reviews and additional information. Convenience aside, it’s an easy way to help consumers create a shopping list or identify products they would otherwise not be able to.

State Farm Insurance partnered with popular car-shopping website Edmunds to develop an app called CarCapture. While it struggled with accuracy early on, the app allows users to take a photo of any car and see general information about that make, model, and year, as well as features and specs, safety specifications, and the “blue book” value of that vehicle. Users can also read consumer reviews, locate nearby dealerships, and request an auto insurance quote or loan estimate.

My personal favorite is Fetch, a Microsoft-developed AI application that allows you to identify the breed of any dog simply by taking a photo of it. Now, if only it were able to show you a list of all similar adoptable dogs available in your area! Someday.

AI-based martech will change the way we think about customer behavior.

Salesforce is all-in on the idea of an AI-based customer sales platform, and other players in the sales and marketing technology space are making similar strides. Not only can machine learning revolutionize the way we qualify, track, and communicate with leads, it can also help us collect and interpret consumer behavior data in ways that we never thought possible, even to the point of making scarily accurate predictions about future behavior.

There’s even an AI-based platform for influencer marketing. Influential uses visual and personality APIs to match celebrities and high-profile influencers to brands and campaigns based on what type of market they are most likely to attract. Imagine using artificial intelligence to match the perfect Snapchat star to your product's campaign though some of us would settle for an AI that could simply explain to us how Snapchat works. Am I supposed to tap or swipe? Can I swipe in that direction? Who knew such a simple app could be so confusing?

How will these opportunities affect your business?

Expect all sorts of novel applications of AI for both consumer- and business-facing brands. We’re still in the early days of machine learning, but we’re already seeing more streamlined processes and workflows, better use of real-time data, and more efficient ways to interact with and serve customers. If your business isn’t already exploring ways to implement AI into your digital marketing efforts, you may find yourself playing catch-up before you know it.