Short can be powerful. Just ask Muggsy Bogues, who scored nearly 7,000 points during his 13 years in the NBA and led the 1986 U.S. National Team in steals and assists — despite being just 5’3”.
Short content can make a similar impact. After all, it’s the short stuff that hooks you. The subject line that compels you to ignore the rest of your inbox, the article headline that grabs your attention over morning coffee, the landing page CTA that inspires you to hand over your email address, the social media post that convinces you to stop scrolling — they’ll help kick your content marketing efforts into high-gear. However, grabbing attention, conveying value, and evoking emotion are hard to do in a split second (and often with a character limit to boot). For many content marketers, the short stuff actually takes the longest to create.
Our Firebrand Labs team is experimenting with all kinds of generative AI tools, spanning content marketing, growth marketing, and PR. I’ve had tons of conversations with the robots, and in the process, I’ve come to a big conclusion: generative AI’s content marketing value is in the short stuff.
Going long and short with generative AI tools
I’ve tried to do it the other way, starting with a topic, some key points, and some brand voice details, and asking generative AI tools like ChatGPT to create a blog post from scratch. Each time, there have been several significant problems:
- The brand voice never shines though, no matter how many descriptive details I provide.
- There is an overuse of cliches that would give my seventh-grade English teacher palpitations. As for me, my eyes might roll right out of my head if I see one more sentence that starts with “In today’s world” or “In today’s landscape.”
- Fact-checking takes as much time as doing pre-writing research.
- Best practices, like short paragraphs and simple language, often aren’t followed. Who wouldn’t hit the Back button if they saw a wall of text that included phrases like “the organization’s ethos of sculpting, deploying, and amplifying extraordinary software”? (And yes, that’s actually something ChatGPT-4 generated for me.)
- It takes a lot of prompts to get the tool on the right path, which is very tedious and time-consuming.
My generative AI aha! moment
I love a good simile, so here you go: using generative AI in content marketing is like resizing an image. If you try to make the finished product really big, it gets pixelated and distorted. But if you make the finished product smaller, the image’s original integrity remains intact.
My results are much better when I take something longer, like a blog post, and ask generative AI tools like ChatGPT to come up with five title options or social post options for it. The same goes for emails and subject lines, and even ebooks and titles in platforms where the character limit will allow.
Does it generate fabulous results every time? No, but even if the output isn’t quite right, it opens up new avenues for brainstorming in a matter of seconds. This is helpful because I have always come up with several title options to make for a smoother editing process. It’s much easier to write a list of choices while I’m in my creative flow than it does days or weeks later if I’m asked for an alternative.
See my generative AI conclusion in action
I started with the tool everyone is talking about: ChatGPT-4. First, I asked it to write five social posts I could use to promote this blog post:
Post 1: 📈 55% of B2B buyers rely more on content for purchasing decisions in 2022. Stay ahead with a content marketing audit! #B2BMarketing #ContentAudit
Post 2: 🎯 A B2B content audit helps you analyze your buyer’s journey, identify gaps, and create a strategic plan for success. #ContentStrategy #BuyersJourney
Post 3: 🚀 Boost your content marketing by finding out what channels, topics, and formats resonate best with your target audience through a B2B content audit. #ContentMarketing #B2BAudit
Post 4: ⏰ Are you spending your time wisely? Discover how a content marketing audit can help optimize your efforts and increase engagement rates. #TimeManagement #ContentAudit
Post 5: 👥 What’s it like to be one of your leads? Improve their experience with a content audit that uncovers critical opportunities. #LeadNurturing #B2BContent
Not bad, right? I like that it included emojis and hashtags. To be honest, I wasn’t expecting ChatGPT’s copy to be that “human.” I then asked it for five more options, but to make them more creative:
Post 1: 🎯💡 B2B content audit: Your treasure map to uncover hidden gems in your content marketing journey. Time to set sail! ⛵ #ContentTreasures #B2BMarketing
Post 2: 🔮 A B2B content audit is like a crystal ball – gaze into it, and find the answers to your content marketing mysteries. #ContentMagic #FutureSuccess
Post 3: 🌱 Cultivate a thriving content marketing garden with a B2B content audit. Nurture your leads, and watch your business grow! 🌻 #LeadGardening #ContentBloom
Post 4: 🎢 Don’t let your content marketing efforts be a rollercoaster ride! Steady the course with a B2B content audit and enjoy a smoother journey. #ContentCoaster #AuditAdventure
Post 5: 🔎 Channel your inner Sherlock Holmes with a B2B content audit! Investigate your content marketing performance and crack the case of business success. 🕵️ #ContentDetective #B2BMystery
Yes, it’s a little off the wall.
Next, I asked ChatGPT-4 to give me five ebook title options based on a very detailed outline I created:
Title 1: “The Full-Funnel Content Strategy: Navigating the B2B Buyer Journey”
Title 2: “Mastering the Content Maturity Matrix for High-Impact Marketing”
Title 3: “The Power of a Well-Oiled Content Engine: A Guide to Marketing Success”
Title 4: “Multiplying the Impact: How High-Impact Content Marketing Complements PR, SEO, and Paid Programs”
Title 5: “Starting the Path to High-Impact Content Marketing: A Comprehensive Guide”
Spoiler alert: I won’t be using any of these, but they did provide some helpful brainstorming fodder.
After that, I moved to another popular generative AI tool: Copy.ai. For an apples-to-apples comparison to the ChatGPT copy, I asked it to write five social posts to promote the same blog post:
Post 1: Ready to take your B2B content marketing to the next level? Check out our comprehensive audit guide and maximize your strategy!
Post 2: Is your B2B content marketing underperforming? Learn how a thorough audit can help you unlock its full potential in our latest blog post.
Post 3: From funnel mapping to user experience, our latest blog post covers everything you need to know about conducting a successful B2B content marketing audit.
Post 4: Don’t let your B2B content marketing go stale. Learn how to refresh your strategy with a thorough audit in our latest post.
Post 5: Discover the benefits of a B2B content marketing audit and how it can help you improve engagement, ROI, and user experience in our latest blog post.
Again, not bad. They’re not quite as “human” as ChatGPT’s writing, but they’re a decent starting point that I can add some oomph to.
I wanted to see how Copy.ai handled titles, so I asked it to write five creative options for the same blog post:
Title 1: “Maximizing Your B2B Content Marketing Strategy: A Comprehensive Audit Guide”
Title 2: “From Funnel to User Experience: The Power of a B2B Content Marketing Audit”
Title 3: “Revitalize Your B2B Content Marketing with an Expert Audit”
Title 4: “Unlocking the Full Potential of Your B2B Content Marketing: The Audit Solution”
Title 5: “Get Ahead of the Game with a B2B Content Marketing Audit: Your Ultimate Guide”
In my opinion, the quality of these was just as high as ChatGPT’s titles, although I don’t think they created the same amount of intrigue that the actual title of the post did (that I created without generative AI’s help). But I did like that it created copy from a few different angles because that helps create a stronger brainstorming session.
Since I was on a roll, I decided to switch things up. I asked Copy.ai to write a blog post about the benefits of a B2B content marketing audit that includes mapping assets to their location in the funnel, analyzing key engagement metrics, assessing email marketing campaigns, and analyzing the user experience on the website:
What is a B2B Content Marketing Audit?
A B2B content marketing audit is a process that helps you to identify and prioritize opportunities for improving your content marketing strategy. It involves analyzing current performance, identifying gaps in performance, and making recommendations for improvement. The goal of an audit is to provide actionable insights into how you can improve your business through better use of content marketing tactics like blogging or social media campaigns.
Mapping Assets to their Location in the Funnel
The next step is to map your assets to their location in the funnel. This will help you identify gaps and optimize existing assets. The user journey is an effective way of understanding how customers interact with your brand, products or services, and how they move through each stage of their buying process. The goal here is not just to understand where they are on their journey but also why they’re there–and what else could be done to help them move forward?
Analyzing Key Engagement Metrics
The first step to a successful content marketing audit is to analyze key engagement metrics, such as:
Tracking engagement. What is the average time spent on your website? How many visitors click through from social media channels? Are you seeing a lot of repeat visits or are people coming in one time and then leaving, never to return again? These are all important questions that will help you understand what needs improvement on your site.
Understanding audience insights. How old are your customers, where do they live and work (geographically), what do they like/dislike about their current experience with B2B products/services? The answers to these questions will give you insight into what kind of content resonates most with this specific group of people so that when creating new pieces of content for them in the future there’s no guessing involved–you’ll know exactly what works!
Identifying trends over time: As mentioned above, it’s important not only knowing how many times someone has visited but also how long each visit lasted; how often did they come back after visiting once before; did anything change between visits (such as new features being added)?
Oh boy, here we go again. It’s the big pixelated JPEG, so to speak. It’s repetitive in some spots, deficient in others (where’s the part about email marketing?), and nowhere near a level of sophistication I would ever consider publishing. And I’ve never gotten anything better from ChatGPT, so I’d say they’re evenly matched (albeit at a much lower standard). From past experiments, I know it would take many, many prompts to get this anywhere close to passable, so I stopped here.
The final takeaway
While the tools are more helpful than I would have guessed a couple of months ago, generative AI is not the content marketing magic bullet many people want it to be, at least not right now. And quite honestly, you’ll never be able to remove the human element altogether. But getting a different point of view — human or robot — on something that has to be as eye-popping as short content is a good thing.
Nicole Pytel is Vice President of Content Marketing at Firebrand Communications. With 15 years of content marketing and branding experience, she loves combining multimedia creativity with strategic data analysis to help clients reach their biggest goals.