Your Next Marketing Hire Shouldn’t Be a Prompt (Unless You Hate Results)
Most people are using AI to do mediocre work faster. The smart ones are using it to do better work differently.
AI in marketing mostly sucks. It floods the internet with soulless sludge, rewrites originality into oatmeal, and convinces lazy teams they’re “innovating” by regurgitating prompts.
But - beneath the noise and fake guru hype there’s a sliver of something real. When used right, AI doesn’t replace your brain. It frees it. It doesn’t generate genius. It accelerates judgment.
Most people are using AI to do mediocre work faster. The smart ones are using it to do better work differently.
Why AI sucks in marketing
The current state of AI in marketing is like watching someone try to microwave a steak and pretend it’s fine dining. Everyone’s automating “creative work” and somehow ending up with less creativity than ever. Teams are feeding prompts into LLMs, getting beige soup in return, and calling it innovation.
Marketers keep trying to use AI to sound “human,” which is adorable in a deeply concerning way. The output? Content that reads like it was generated by a tired intern who just discovered synonyms. And let’s not forget the design teams slapping AI into brand campaigns like it’s glitter glue. Just because the visuals are technically generated doesn’t mean they make sense - or match your brand, or look good, or do literally anything except exist.
We’ve also entered the era of AI goop. The internet is drowning in slop: AI-written articles, AI-voiced videos, AI-generated shorts, AI-created podcasts. It’s mass production without direction. Recruiters are hiring for “prompt engineering,” which is code for “please make the robot slightly less embarrassing.”
The robot isn’t your strategist. Stop pretending it is.
And in case you were wondering, yes - Google hates it too. Because when your AI blog post has all the depth of a puddle and none of the soul, Google doesn’t want to rank it. Shocking, I know.
The TL;DR? We’re using AI to fake value at scale. And it shows.
AI helped me save time, so I could ignore worse content faster?
Also AI gives me more time. Cool. But that doesn’t magically make my time valuable.
Here’s a genuine question I’ve been asking myself:
Who’s out here willingly listening to AI-generated podcasts? Like… unironically?
Because unless you’ve completely given up on brain activity for the day, I can’t see the appeal. These things aren’t conversations. They’re just verbal white noise. Background static with a slightly smarter accent.
It’s not even podcasting at this point. It’s audio doomscrolling. You’re not listening to learn anything (and I mainly spend time in this department to do that). You’re not engaged. You’re just passively letting synthetic voices spoon-feed you trivia until your will to live quietly exits the chat.
And yeah, I get it - sometimes you want something mindless. But maybe, just maybe, we don’t need to industrialize content emptiness.
Maybe we don’t need 10,000 AI podcasts reading Reddit threads and calling it insight.
So I’m asking myself: if I’ve got time, why spend it like that? What am I actually gaining by tuning in to a conversation that no one ever really had?
Spoiler: nothing.
But wait, we still have something good happening!
Where AI actually benefits marketing
Alright, credit where it’s due: AI isn’t all noise.
Occasionally, it does something useful. Like handling the boring stuff marketers pretend to love - spreadsheets, campaign reports, 47-tab competitor audits. Turns out, robots are pretty decent at busywork.
Used right, AI is less “creative partner” and more “intern that doesn’t talk back.” It drafts, analyzes, outlines, and crunches numbers without needing snacks or a motivational Slack channel.
The real marketers - the ones with a brain - use it to speed up grunt work so they can focus on strategy, storytelling, and not hating their jobs.
Some tools are even helping with campaign planning and SEO optimization. Fancy that. Real-time ad testing, content performance prediction, deep market scans… AI can stack up insights faster than your 17-person strategy team ever did. Just don’t expect it to come up with a tagline that doesn’t sound like it belongs on a shampoo bottle.
Bottom line: AI works best when it’s invisible.
It doesn’t replace your job - it just removes the part that made you want to rage-quit.
Where AI still sucks (for me)
Let’s talk about the “intelligence” part of artificial intelligence.
Because half the time it feels like it’s just confidently wrong. You ask it for research, and it throws out links that go nowhere. Literal 404s like it’s time-traveling back to the early internet.
And if you’re hoping it’ll match your tone of voice? Good luck. Unless your tone is “corporate LinkedIn post written by a mildly caffeinated someone,” it’s going to miss by a mile.
Even when it gets the facts mostly right, it’s surface-level fluff unless you’re feeding it prompts that look like a legal brief.
Speed? Sure. Quality? Not unless you micromanage it like it’s an unpaid intern who just lied on their resume. And every now and then it just… makes stuff up.
Full hallucination mode. Like, thank you for the confidence, but no, Shakespeare did not invent TikTok.
Where it actually shines (personal insights)
That said—when AI does work, it’s kind of wild.
As a brainstorming partner, it’s a solid 9 out of 10. Not because it has genius ideas, but because it forces me to think faster and weirder.
But the trick is, you need a framework. Otherwise, it’s like pouring gasoline on a sandbox and calling it fire.
It’s also secretly my English coach. I’m not a native speaker, so having something clean up phrasing or tighten structure without dumbing things down? Huge win.
And when I’m neck-deep in strategic work, AI gives me structure without slowing me down. It’s like having a second brain that’s good at outlining but bad at vibes.
Promo campaign ideas? It spits out 20 in 10 seconds. Most are trash. Some are gold. But that’s the game.
Also? It’s my Martech therapist. When I’m about to throw HubSpot out the window or break up with Zapier for the fifth time, AI helps me troubleshoot before I start crying into a Google Sheet. Not glamorous, but very real.
The “don’t get slop” framework for brainstorming with AI
(aka how to use a robot without letting it waste your time)
AI can help you go deep - but only if you stop treating it like a magic 8-ball and start treating it like a mildly competent intern with memory issues.
Here’s how I actually use it for meaningful, structured research and creative ideation.
1. Split your brain dump into small, painful chunks
Bad: “Help me build a go-to-market strategy.”
Better: “Let’s start by mapping target audiences. Then we’ll look at positioning. Then channels. Then funnel stages.”
Why it works: One giant prompt = one giant mess. Break big questions into tiny steps. More back-and-forth, yes. But also more signal, less sludge. Think of it like a slow roast - not microwave content.
2. Start with context, not a question
Bad: “What are the top marketing trends?”
Better: “You’re a growth strategist helping a B2B SaaS founder go from 0 to 500k ARR. What specific trends in paid acquisition are worth testing in 2025?”
Why it works: AI isn’t smart. It’s predictive. Give it a job, a role, and a scene to work in - or you’ll get advice that sounds like it came from a business horoscope.
3. Always ask for categories before answers
Bad: “Give me 10 campaign ideas”
Better: “Before we get ideas, give me 4 angles we can explore: audience pain, timing, incentives, and channels. Then ideate under each.”
Why it works: You want range and focus. Asking for buckets first stops the rinse-and-repeat answers and forces more thought per output.
4. Use competitors to anchor reality
Before asking for ideas do some deep-dive:
“Based on [competitor A, B, C], what are the common patterns in their campaigns? What are they missing that we could punch through?”
Why it works: AI works better with reference points. Giving it live market context keeps your ideas grounded - and sometimes even helps you zig where others are zagging.
5. Double tap with: “What would a contrarian say?”
After any list:
“Now challenge that - what’s missing, outdated, or plain wrong?”
Why it works: AI is a people-pleaser. Make it debate itself. You’ll uncover blind spots and sharper insights that feel like they came from a real strategist, not a content mill.
6. Ask for source logic, not just output
“Explain how you got this. What’s it based on - behavioral trends, case studies, assumptions?” (AKA give me proofs!)
Why it works: If it can’t justify the take, it probably hallucinated it into existence. And if that’s your idea of ‘research,’ you’re gonna have a bad time.
7. Summarize, then remix
Wrap it with:
“Summarize the top 3 insights. Then give me a weirder, riskier version of each.”
Why it works: Forces clarity, then creativity. You get one version you can ship and one that might actually be interesting.
final boss: use your brain
No matter how smart the prompt looks—don’t copy-paste your way into irrelevance. AI still hallucinates, misses nuance, and occasionally makes stuff up with the confidence of a guy who’s never been punched in a meeting.
Cross-check. Think critically. Own the take.
Otherwise, you’re just dressing up noise and calling it strategy.
Your next marketing hire will not be a prompt
“Your next marketing hire will be a prompt.” Cute line. Great for pitch decks. Still absolute nonsense.
It sounds bold until you realize it’s just tech guys cosplaying as marketers. A prompt isn’t a hire. It’s a sentence. It won’t catch a weak insight, question a bad brief, or tell you your funnel’s a mess.
A good marketer doesn’t just ask the tool for ideas. They shape the input, question the output, and turn advise into execution. That’s not a prompt. That’s judgment. And no matter how good AI gets, judgment doesn’t scale.
So no - your next hire shouldn’t be a prompt. It should be a human who knows when to hit backspace.