Can AI improve your marketing and help grow your business? Contributor Nick Wassenberg whips up some prompts and explores.
It’s all over the news. Products such as DALL-E 2, ChatGPT, MidJourney, Jasper, Copilot and more have heated up, and big brand software companies have layered on more usable tools such as Solvvy from Zoom and Muse AI from Google. By the time you read this, I’d be surprised if there weren’t many newer tools in the headlines. For marketers, there’s some fear. I get it. I don’t want to get left behind. I’ve been in tech marketing most of my career, and I’ve seen two trends change everything.
First, the rise of internet as the primary information source. My first office job was at an insurance agency in Minneapolis. When I started, we had one computer with a dial-up modem for completing auto insurance quotes through progressive.com (sounds prehistoric). Just a couple years later, I was building content marketing and lead-gen websites for our niche insurance projects. That’s right, you may remember the classic Union Trustee Fiduciary Liability Insurance Resource Center website. Before its time, I’m telling ya!
Jokes aside, the information—and power—dynamic shifted right under our noses and has accelerated. Hoarding information and winning through “information asymmetry” (as Dan Pink put it) was no longer going to work.
The second shift was automating the process of buying. I’m a marketer, not a salesperson. But think about how much of purchasing has changed. In the span of only a couple decades, e-commerce went from nearly $0 to $1 trillion in gross merchandise value in the United States alone. That’s a big shift. But even beyond the shopping cart for clothes, electronics, or household goods, the biggest part of buying is also done primarily online—evaluating options. Buyers look to Google, Amazon, Facebook, Instagram, Twitter, Reddit, YouTube, and TikTok for product reviews and new things to make their life better. Add G2 or Capterra specifically for business software. Sites such as Nerdwallet, Yelp, IMDB, Rate My Professors, and so many more have built valuable businesses creating new ways for buyers to access information.
This shift for marketers was big. It meant rethinking interruption-based advertising such as direct mail, out-of-home, and other ad channels. Marketers effectively moved to automating the buyer’s journey through marketing automation—contextual “in-the-moment” digital advertising. They also took advantage of a flood of new data for understanding audiences and built interactive experiences like smart(ish) chatbots. Social media and content marketing also went beyond branding to pinpointing messages on awareness, consideration, and purchase paths.
Is artificial intelligence (AI) the third wave of marketing change I’ll experience? I think it could be. It directly puts information in people’s hands instead of in a search engine or gatekeeper. It scales the synthesis of information in ways never possible before. It’s not without its dangers and unknowns, but the potential is massive.
The good news is that the fundamentals of marketing (understanding behavior, crafting compelling messaging, and effectively matching the needs and wants of people to a distributed offer) have never been more important. My belief is that AI will accelerate successful marketing. So, marketing folks, don’t be afraid. Be prepared.
Understanding the Basics of AI
We’ll get into more depth in a second, but I can’t resist having ChatGPT give us a definition:
Prompt: What is artificial intelligence?
Artificial intelligence (AI) is the ability of a computer or machine to perform tasks that typically require human-like intelligence, such as understanding language, recognizing images, making decisions, and solving problems. There are various types of AI, including narrow or weak AI, which is designed to perform a specific task, and general or strong AI, which is designed to perform a wide range of tasks. AI has the potential to transform many industries and has already been applied in a variety of fields, including healthcare, finance, education, and transportation.
Not bad, little buddy. That’s simple of course. Marketing people need to go deeper. There are ample resources to learn the different levels of AI, hear from experts where AI is going next, and get smarter about AI in the context of modern marketing. It’s also important to understand the potential downsides, too.
Learn the basics first, and then get experience. As with a lot of technology, the best way to learn is to try it out. I’m closest to my daily work at Fulcrum, so that’s where I started. Fulcrum provides a software system for small to medium-sized manufacturers in the United States. The company has multiple successful ways to market the company message and generate pipeline of potential B2B buyers, but there are hundreds of thousands of potential fits, and we’re really just getting started.
What happens when we give AI a few marketing assignments? Let’s find out.
Exploring AI as a Marketing Tool
Prompt: Create a landing page tagline
Our product is evolving, improving, and expanding every week (a big part of how we differentiate from stagnating enterprise resource planning (ERP) systems), so creating compelling website copy as we grow is vital. Can AI help? Let’s see.
Prompt: Create text for a social media ad.
Using Jasper.ai, we entered some simple text about an upcoming livestream to get ad copy for LinkedIn and Facebook.
Prompt: Design a company mascot.
We have an incredibly talented group of designers and creative marketing people. Sometimes this combination sparks semi-off-the-wall ideas. I love it. Here’s a recent one. What if our mascot was a hardworking, charming raccoon?
Running a few simple prompts through MidJourney provided some fun results (this is a small sample).
We didn’t find a slam dunk mascot through a few MidJourney prompts, but it did accelerate some thinking in terms of visuals. Where we take it from here is up to us.
Prompt: Describe customers’ problems.
Unless we understand the people we want to reach and engage, any successful marketing campaign is just luck. I try to eliminate any arrogance we might have in believing we have our audience all figured out. Can AI help us build and maintain a better understanding?
Let’s look at a very simple prompt and result through ChatGPT.
This response is impressive, accurate, informed, and dangerously insufficient. It’s perhaps a starting point for more research, but in my experience the details about what potential customers say—the words they use, the pains that are always top of mind, and needs behind the needs—are the foundation of great marketing. Without more depth, this simplified and summarized AI post is a path to mediocrity.
Prompt: Create outbound messages.
For early-stage companies such as Fulcrum, a potential client’s first awareness can come from outbound messages. Getting the messaging right is important for generating pipeline and getting us on the radar for prospects. Bad, generic, annoying, non-differentiated messaging for outbound is dead. We don’t do it and I personally don’t respond to it. Can AI help us do this difficult task?
ChatGPT again gives us some answers to some basic prompts.
Each one of these would get instantly deleted by me as a recipient. There’s the lesson—do something that looks nothing like these. That could mean creative content that engages buyers before a sales development representative reaches out and getting even more personal and contextual with outbound messages. AI can help the team avoid the generic by using it as an anti-pattern. If it looks AI-generated, it’s not good enough.
What We Learned
The examples above are for illustration only. There’s room to improve the prompts and go deeper, but the examples do show getting better at understanding an audience in order to write effective prompts is an area to focus on.
Using AI for brainstorming help is a no-brainer use case (sorry) but it’s only as good as the actual understanding the marketing team has in the first place. There is no replacement for listening to customers.
As others have pointed out, some areas of marketing - I’d say especially B2B tech marketing - have stagnated as they’ve tried to play the game Google and others have designed. Keyword stuffing in a blog, unhelpful gated content behind copycat landing pages, and artificial case studies are the past not the future.
Creative application of intelligence is how marketing wins. How much of that creativity will be helped by artificial intelligence is up to us to find out.