In this episode of Marketing Roundtable, we explore various use cases for AI–and more specifically, large language models (LLMs)–in the workflow of our marketing agency.
We had the pleasure of speaking with Project Manager, Ben Rawner; Web Developer, Andrew Okupinski; and VP of SEO, Ron Sansone to see how they use LLMs in their daily work.
The takeaway: Artificial intelligence is capable of wearing many, many hats.
Let’s dive in.
Ben Rawner, a project manager here at BrainDo, has been a heavy adopter of AI in his workstream. And he’s always looking for new ways to refine his processes.
A project manager’s responsibilities typically call for great communication skills, concise documentation, and the ability to capture detailed notes. Ben has created a system with AI to find efficiencies in all of the above.
Day to day, Ben juggles meetings with multiple clients and team members on many different projects. Taking detailed and relevant notes can be a daunting task, but AI can make that process easier.
Ben currently employs the following note generating process:
Ben shares that right now the note generating process is not ideal, but he believes that AI is very close to a breakthrough on this front.
With his well-tested and optimized prompts ready to go, he can provide actionable next steps to clients quickly.
Prompts are a theme throughout this blog: Good prompts produce good results.
Tip: Once you have nailed down a prompt that yields the desired results, save it! Create a prompt library for you and your teammates to use over and over.
When writing, Ben uses AI to suggest better grammar, adjust tone, and use concise structure. While Ben is always the primary writer of his correspondence and document, he likes to use AI as his personal proofreader.
"You know, sometimes it's just nice to have another set of eyes on it," acknowledges Ben. "I think [AI] is really great for that."
Tip: When writing a business case for why a client should do something, ask an LLM to provide reasons why a client should not do it. It can play devil’s advocate so you can construct an air-tight case.
Andrew Okupinski, a long-time web developer at BrainDo, uses AI to help with troubleshooting.
Before the wide accessibility of ChatGPT, web developers used Google to find possible solutions for broken code–referencing StackOverflow answers or reading through tedious documentation. At times it felt like a needle in a haystack.
ChatGPT acts as a centralized source for finding solutions, bypassing hours of searching and trial-and-error. However, AI responses can be hit or miss depending on the prompt, search, and terminology used to get the answer.
With the correct prompt, ChatGPT can even be a great resource for expanding on current knowledge and learning new languages.
Tip: You can write code in one language (most likely the one you are most familiar with) and ask ChatGPT to translate it to another.
"It's super helpful if you're just learning coding or you're learning a new language but you have that background in another one."
And remember our rule: Good prompts produce good results.
This mantra is echoed by our next guest, our content connoisseur, Ron Sansone.
To call Ron Sansone, BrainDo’s VP of SEO, an AI savant is an understatement.
Throughout his personal journey of exploring various AI capabilities, he has tested over 70 different models and counting. He has been able to implement AI into his day-to-day processes with continuous refinement and experimentation.
With this experience on hand, Ron gives his advice for crafting an effective AI prompt.
"The prompt creation is probably the most important part," he shares. "This might be counterintuitive, but you might actually need to write a little bit more to get the right kind of response out of an AI tool, like ChatGPT or Gemini, than you think you would."
Tip: Write more to write less! Having a more detailed prompt with lots of context will likely provide a better result.
Here’s the full breakdown of Ron’s prompt-crafting advice:
For SEO specifically. Ron shares three things that AI is able to assist with from an SEO perspective: analysis, ideation, and summarization.
For the analysis and ideation side of AI, Ron has used ChatGPT to parse through Google Console data to assist with finding search trends. It is able to analyze the data, identify emerging search query trends, and provide new ideas from it. This methodology helps create net-new content, rather than producing skyscraper content for organic search ranking.
Various AI models can also be used to interpret the content of images, which is great for creating a series of Alt tags for multiple images in an article. Ron notes that some results in this ideation phase can be very creative and are a great starting point upon which to expand. Cue our obligatory reminder that this requires a good prompt to produce good results.
Ron also uses AI to summarize these analysis documents into executive summaries and next steps, similar to Ben’s methodologies. This is particularly useful for communicating technical SEO findings that may be challenging to concisely explain.
As you can see, AI can wear many hats. While we’ve focused on the three disciplines above within this article, the full Marketing Roundtable podcast also discusses AI use cases for design and organic social media–with a little help from some ChatGPT 4o responses.
No matter the discipline or role, there are many ways to implement AI into our daily tasks. So give it a try if you haven’t already.
For the full discussion, check out the Marketing Roundtable podcast episode in video or audio format.
The Marketing Roundtable is available to watch on YouTube or listen to on every major podcast platform.
Need help with building out digital marketing campaigns? The BrainDo team is always ready to collaborate, from management to competitive analysis. Reach out to BrainDo today!