How to Get Started with Artificial Intelligence in Inbound Marketing

INBOUND 2018 HubSpot Overit Marketing

This afternoon we heard from PR 20/20 CEO Paul Roetzer.

Artificial intelligence (AI) is a pretty heady subject, so I want to start with two disclaimers:

  1. The content of this talk and this blog are meant to be more high level – what’s happening with AI, where it’s trending, and how it’s being integrated into marketing and sales.
  2. AI is still in the early adoption stage. Even though we’ve been watching sci-fi movies with AI for years, most of that AI has not been created yet. We’re still in the relatively early stages of implementation, especially as it applies to the marketing field. So don’t feel like you’re behind, because you’re not!

Let’s start by breaking down what AI is (and what it isn’t).

El Dictionary defines artificial intelligence as “the theory and development of computer systems able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages.”

Whether we realize it or not, we’re seeing AI all over the place – Alexa and Google home. Our shopping carts and suggested purchases on Amazon. The movies Netflix thinks we should watch next. Even the rerouting our Google Maps does to help us avoid late afternoon traffic.

In the marketing world, AI has enormous potential.

Ever think to yourself, “Dang, I am DROWNING IN DATA”? (No, only me?).


Thing is:  most of us only use a fraction of our data; the rest of it is landing in no man’s land and not being mined for insights or potential use cases.

As human beings, we can only process a certain amount of information at a time. I mean, it’s only day 1 of Inbound (Day 2, if you count Partner Day), and I’m practically snoozing in the Partner Workspace trying to absorb it all!

The incredible thing about machines is that they have virtually limitless capacity to learn and adapt. Whereas one individual (or even a team of individuals) can’t possibly process tens or hundreds of pieces of data about thousands or millions of leads and contacts, AI can do it! This is why Mr. Roetzer says the future is Marketer + Machine. AI can’t do everything – there are still lots of roles to play in the marketing world, even with the introduction of AI.

Potential Use Cases for AI

Now, if you’re like me, you’re like, “Yeah, that’s cool and all, but what can it actually do?” Because, let’s be real, at the end of the day it’s about the results an employee (or an artificial one???) can produce.

Here are some potential use cases for AI in your business:

Sample Production Use Cases:

  • Analyze and edit content for grammar, sentiment, tone, and style
  • Convert text to voice
  • Create data-driven content
  • Predict content performance before deployment
  • Tag images and content with keywords and categories << !!!

Sample Personalization Use Cases:

  • Engage in users in conversations through bots and chat
  • Optimize email send time at the individual recipient level

Sample Digital Marketing Use Cases:

  • Adjust digital ad spend in real-time based on performance
  • Identify social media and news trends for promotional opportunities
  • Moderate comments at scale across social channels.

Sample Performance Use Cases:

  • Create performance report narratives based on marketing data
  • Discover insights into top-performing content and campaigns
  • Forecase campaign results based on predictive analysis

Curious what AI can do for YOUR business?

Try this simple exercise from this afternoon’s presentation:

  • Evaluate repetitive, manual marketing tasks that could be intelligently automated.
    • What do you do repeatedly?
    • How much time does it take?
    • How valuable would it be to automate it?
    • Then research AI that can do that thing.
  • Asses opportunities to get more out of your data – discover insights, predict outcomes, and devise strategies. These are more opportunities to implement and leverage AI.
  • Consider the AI capabilities of your existing marketing technology stack. If you’re using a tool such as HubSpot, there are AI capabilities built into the platform. Check out what’s possible within your existing tools to see how you might be able to save yourself some time on things that you’re currently doing manually.

Two great resources for you guys:

(because who doesn’t love good resources?!)

This walks you through use case scenarios and helps you score the value AI might have for you. Super super helpful and easy to use, so check it out.

IBM’s AI (Watson) is doing a lot of cool things. You can actually test out each of the products in their suite to see what the tool is like in use. NEAT-O.


Finally, and I would be remiss to leave this out, as it is crucial for marketers to note, is that there’s a difference between live chat, bots, and AI.

Tools that ask, “Was this helpful?” (think:  ZenDesk’s Satisfaction Predictions) are AI. They are taking in information, learning from it, and changing the output.

Chat bots are based off of branching logic that is generated and implemented by a human, but the tool itself doesn’t actually do any thinking. It is not learning from the interaction, and thus is not considered AI. Interesting, eh?

Well, there you have it. Some of the ins and outs of AI and some of the incredible things it’s capable of doing in your business. Let us know how you’re thinking of using AI in your business!