Practical Tips from my 2-week AI deep Dive- A.K.A Why I wish I had gotten a Communications Degree

Kara Babb
Mar 15
4 min read

The past two weeks I’ve been ripping the band-aid off, ignoring my guilt about global resource depletion and diving head-first into AI. I list a bunch of tools and resources at the end, but I want to sit with an idea with you for a moment.

Me, a Global Business studying, Mandarin-speaking SNOB wish I would have instead studied Communications (or at least minored in it). A degree I always thought was just a way to get through college without doing any real work (a degree in TALKING?! please).

Why? Because the future AI Wizards aren't going to be the ones who know all the models and can code. But instead those who are strong Communicators, who know what to say and how to structure their prompts to get the most out of any LLM or AI tool they interact with.

This idea came to me from The Algorithmic Bridge, stating that you can never underestimate the value of a good prompt.

“Deep Research is really good—if you know how to prompt it. Slight tweaks in your prompt can yield outsized gains in performance. Or the opposite. Those who know the right tweaks will see an incredible return on their investment—currently $200/month.” – Alberto Romero

The article is excellent I recommend reading with no fewer than 2 cups of coffee if Math isn't your thing. In short, what I got out of it was If you are a regular human feeling overwhelmed keeping up with the models and tech-speak, take a breath and instead focus your efforts and energy on improving your communication skills. The best part? These skills can be applied to working with more junior team members, or offshore resources who require VERY specific SOP’s.

As the inimitable Sarah Falcon says, “AI is like a brilliant, but very green intern” If you can clearly communicate what you want you will unlock their brilliance. Also verify their info because they can be very wrong.

Now you might ask, "but Kara how can you actually use this in Ecommerce?" Well I did a pretty awesome deep dive on the female menstrual health category via DeepResearch. It was an extensive conversation but yielded information I either wouldn't have found otherwise (Reddit scrapes, extensive news scrapes, product detail scraping, social media data, etc.) or would have taken me hours to complete.

Here is my entire chat - Best Organic Period Care

I hope the resources below are helpful, now I’m logging off screens because all this talking is giving me a headache.

Here are the tools and functions I’ve learned about these past 2 weeks:

ChatGPT projects & Prompt outline (Image/Idea credit: Kyle Lacey)
New course I am trying: Prompt Engineering | Coursera
Prompt idea: Something else I’ve tried this week in my prompt is outlining how important the research will be, aka what is at stake (example at the end). AI was built by humans, and just like humans it appears to take more care and produce better results when there is 1) a lot at stake, 2) clear oversight and their mistakes will be caught.
Steal my DeepResearch prompt: "Do an analysis of the [𝗢𝗿𝗴𝗮𝗻𝗶𝗰 𝗠𝗲𝗻𝘀𝘁𝗿𝘂𝗮𝗹 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝘀] Industry. Create a list of the top [𝟱] brands who sell menstrual products. Analyze every aspect you can find from team size and roles, brand strength, prices, and revenue. Also list strengths and weaknesses of each menstrual care brands. Be as detailed as possible. This report will be used by this [𝗢𝗿𝗴𝗮𝗻𝗶𝗰 𝗠𝗲𝗻𝘀𝘁𝗿𝘂𝗮𝗹 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝘀] company to guide their business strategy and millions of dollars are at stake. Scope: All kinds of at [𝗢𝗿𝗴𝗮𝗻𝗶𝗰 𝗠𝗲𝗻𝘀𝘁𝗿𝘂𝗮𝗹 𝗽𝗿𝗼𝗱𝘂𝗰𝘁𝘀]. Rank them based on list size and estimated revenue. Metrics: Dive into other metrics like audience demographics, engagement rates, monetization strategies, partnerships, and so on. Geographical Focus: Global but especially the [𝗨𝗦 𝗺𝗮𝗿𝗸𝗲𝘁] Competitor Benchmarking: Analyze them separately and at the end create a comparison chart. Do you have any questions?"
Steal one that's even better from Annabella Mingo
Searchable prompt library: GitHub - f/awesome-chatgpt-prompts: This repo includes ChatGPT prompt curation to use ChatGPT and other LLM tools better.