30 days into exploring AI for marketers: Ready for some use cases, but not all

A month ago, I reached out to this community for help as I embarked on my journey into the world of generative AI tools for marketing, promising to share my learnings in return. You helped and now it’s my turn to return the favor. Over the past month, I have spoken to 7 experts and practitioners, been hands-on with 11 tools, and worked on at least 20 different assets/projects. What you will read here was my experience. My request as you read this is to let me know if you have found workarounds or solutions to any of the challenges I call out here as I would love to expand my knowledge and proficiency with these tools.

1. Gen AI can produce short form content (less than 300 words) with ease. . . Short form content, such as social posts, emails, and call scripts, is a great way to reach a large audience quickly and easily. It's also easy for AI to write. I have found that most solutions get the content 80% there in seconds, allowing an SME to spend a few minutes of editing time to get the asset over the line. I see this as a great use case for improving the creativity of social posts and nurture emails, as well as bringing a summary box to the top of all long form content (great example here). Tools: Bard, DuetAI for Google Workspace, Jasper, Anyword, and Writer
2. . . . While long form content (more than 300 words) needs a strong SME partner. Long form content, such as blog posts and white papers, can be supported by Gen AI tools, but I have not had success with great outputs when tools are left to their own devices - they really can (and do) make things up (hallucinate as we say in the industry). Instead, use these tools to help brainstorm ideas and write outlines for longer form content - you can even have the tool expand on their outlines section by section - but I have found keeping each output to less than 300 words will garner the best results. The same goes for asking tools to summarize content for you, especially if you are looking for the summary to be a derivative asset like a blog - do it in small chunks and by someone who knows the content. Tools: Bard, DuetAI for Google Workspace, Jasper, Anyword, and Writer
3. Audio-enabled content should be table stakes with Gen AI. Audio is the next big thing in content marketing as it’s a great way to reach a wider audience and engage with people who are visually impaired or are on the go. While delivering high-quality audio versions of blogs and whitepapers used to be a time consuming task, as it required a person to read each asset, new technologies deliver human-sounding audio versions of your content instantaneously. Tools: ElevenLabs
4. AI-generated images have a “wow” factor, but are not useful due to brand requirements. Image creation is no doubt one of the biggest challenges we have as marketers - we always need more (for things like social posts, whitepapers, newsletters, etc). I have played with a few of the tools available and while some of the outputs are really slick, some are, well, not. Some of the weird things I have seen include: a woman’s head on backwards, when uploading a photo of a woman for a new headshot, it came back as a man, and the most common, people looking cartoonish. More than this, I simply can’t get an output close enough to our brand guidelines. If anyone has suggestions on ways they have seen success with image creation (or design in general), I would be very interested. Tools: Imagen, Midjourney, Runway, and Lexica
5. Gen AI puts market and industry research at your fingertips. I have only touched the surface here and will experiment more next week, but this post by Nicole Leffer (follow her!) had me very intrigued. I copied her prompt into two different tools and was quite impressed with the results. The output provided will give teams great insights in an industry or market, which in turn will enable even better content. Tools: Bard and Poe.

Over the next few weeks, I will be further diving into market research, experimenting with solutions for data analysis, and playing more with design tools to help teams produce polished (read: pretty) deliverables. If you have tips or solutions I should look at, please let me know.

My final thoughts: there is no doubt in my mind that these technologies are going to supercharge marketers in ways we never dreamed. That said, these solutions are still in their infancy and need a lot of hand holding to produce content that is ready for publication (and on brand). This means that you must have someone proof-read *everything* generated by these solutions before you publish.?

I hope this was helpful. If you have any questions, please feel free to reach out. And if you have discovered any incredible technologies or use cases, please share - I would love to chat.?

P.S. I want to thank Jeremiah Owyang , Paul Roetzer , Nicole Leffer , Abhishek Ratna , Louise H. , Reto Meier , and Guillaume Roques for sharing their insights with me as I kicked off this journey. If you are looking for really smart people, doing incredible things with Gen AI, these are your peeps!

This article was originally posted on the GenEdge Resources page.

_______________

About the author

Natalie Lambert, Founder & Managing Partner at GenEdge Consulting, is a leader driving innovation in marketing through generative AI. Her journey into the world of AI began at Google, where she initiated AI pilot projects across the organization to identify practical use cases, tools, and strategies to enhance Google's marketing efforts.?Natalie also led the content strategy at Google Cloud, held CMO positions in two successful enterprise startups, and worked at Citrix in various marketing capacities. Her career began at Forrester Research, where she advised companies on tech investments and best practices.