Full Steam 03: The AI gut check for content marketers
Full Steam is an ongoing conversation about the craft of B2B marketing. In each episode, experts from Iron Horse explore a single topic in depth, giving you practical takeaways and a fresh perspective on the challenges you're navigating every day.
Terrible AI-written content is everywhere. Content marketers are scrambling to prove AI’s worth. And writers are spiraling: questioning em dashes, avoiding “It’s not X, it’s Y” constructions, and sometimes even leaving in typos just to prove a human wrote it.
In this episode of Full Steam, Iron Horse principal content strategist Alex Jonathan Brown sits down with content director Amber Keller for an honest discussion about what it means to be a content marketer in the AI era.
Amber breaks down what’s changed with buyers’ research habits, when to leave some human messiness in your copy, and why she’s still optimistic about content marketers’ ability to keep things fresh.
Featuring
- AI is your new audience. Along with third-party sources, LLMs still read your content. Thorough, well-structured writing is helpful for AI discoverability — as well as for humans researching quick answers.
- Readers still crave a human voice. Beyond that basic research, people will always appreciate interesting stories. There’s value in being unique and voice-y when everything else sounds the same.
- Bad AI writing forces creativity. Today’s writers need to stay nimble. Avoid the AI tells and buzzwords of the moment. Stand out by embracing a little messiness (but maybe not full-on typos).
Should B2B companies write separate content for AI research versus human readers?
It depends on what you want that content to do.
Some content exists to answer explicit questions, like what do you offer, what do you do, or why you’re better than competitors. That content should be written in a way that’s easy for LLMs to summarize accurately. It’ll also be helpful for humans wanting quick information.
The other kind of content exists to build affinity over time. Assets like voice-y articles, newsletters, and LinkedIn posts are valuable for humans who are interested and paying attention, even if they’re not necessarily ready to buy.
Define that goal first, then write the content from there.
What steps should a B2B company take to adopt AI in content production without sacrificing quality?
Start by understanding what AI is doing with your existing content.
Amber recommended pasting a finished piece of content into an app like Claude and asking for a summary. Its answer will tell you whether your writing is actually saying what you think it’s saying, and whether AI is accurately representing your ideas when buyers are researching your category.
She also shared a story about a time AI failed as a production tool. When her team tried to use AI to summarize a set of marketing assets, the output was literal to the point of being useless. AI could only describe a document’s content without discerning its value. To get good quality results from AI, you need to train it on what good quality looks like.
What AEO approach do B2B marketers recommend for getting cited in AI-generated buying guides and comparisons?
Write thoroughly enough that an LLM has something worth citing, and make sure it’s pulling the right information.
Depth matters more than it used to. AI can only summarize what you provide. Clear headers, explicit key takeaways, and direct answers to the questions your audience is asking make it easier for an LLM to match your content to a specific research query.
Catch more Full Steam: