Lesson 001Field guide · AI for Creators
Content Idea Generation
Generate ideas that fit your niche without sounding generic.
You'll learn
- Why AI's content ideas feel generic by default
- The three ingredients that make ideas fit your niche: niche, audience, past winners
- The quantity-then-curate workflow
- The "would my audience actually care?" filter
- How to push a decent idea into one only you could make
01Section
Why AI ideas feel generic
Every creator tries this at least once: open an AI tool, type "give me 10 video ideas," and get back a list that could belong to literally anyone. "Share a day in your life." "Do a Q&A." "React to a trend." Technically ideas. Practically useless.
The tool isn't broken. Asked a generic question, AI produces the average of every creator on the internet — and the average of everyone is interesting to no one. The fix isn't a cleverer trick. It's giving the AI the one thing it can't know: what makes your corner of the internet yours.
The rule this lesson is built on
Generic in, generic out. AI ideas are only as specific as what you feed it.
Checkpoint
02Section
The three ingredients
Before asking for a single idea, hand the AI three things. Together they take about five minutes to write down — and they change everything the AI gives you afterward.
Your niche
Not "fitness" — "strength training for busy parents over 40 who work out at home." The narrower, the better the ideas.
Your audience
Who they are, what they struggle with, what they've already seen a hundred times and are tired of.
Your past winners
Your three to five best-performing pieces, and your best guess at why each one worked.
A reusable setup you can paste before any idea request
Before you suggest anything, here's my situation. My niche: [describe it narrowly]. My audience: [who they are, what they struggle with, what they're tired of seeing]. My best-performing content so far: [3-5 pieces and why I think each worked]. Keep all of this in mind for everything that follows.
Privacy note: Your published posts are public — pasting them is fine. Private DMs, subscriber lists, and analytics tied to real people are not. Describe patterns ("my how-to posts do best") instead of pasting raw data.
Checkpoint
03Section
Ask for twenty, keep two
Here's where AI genuinely changes the ideation game: volume is free. A human brainstorm runs out of steam after six ideas. An AI tool will happily produce twenty, thirty, fifty — and the useful ones are rarely in the first five, because the first five are the obvious ones.
So don't ask for "a good idea." Ask for twenty, expect most to be mediocre, and treat the whole list as raw material. Your job shifts from generating to curating — and curating is where your taste, the thing AI doesn't have, does the work.
Ask for volume: Twenty ideas minimum, grouped by type — how-tos, stories, opinions, behind-the-scenes.
Cut fast: First pass: delete anything you've seen a dozen creators do. Don't deliberate — a maybe is a no.
Keep the sparks: The two or three that made you think "oh, I could actually say something about that" — those are the winners.
Key takeaway
AI supplies the volume. You supply the taste. Never publish an idea just because the AI suggested it.
Checkpoint
04Section
The filter: would my audience actually care?
Every surviving idea gets one question before it earns a slot in your plans: would my audience actually care about this — or does it just sound like content? An idea can be clever, trendy, and fun to make, and still be for nobody you serve.
And when an idea passes the filter but still feels a little flat, don't discard it — push it. Three moves reliably turn a decent idea into one only you could make:
Add a constraint
"...in under 60 seconds," "...using only free tools," "...for someone with 15 minutes a day." Constraints create angles.
Add your angle
What do you believe about this topic that most creators in your niche don't say out loud? Lead with that.
Add your experience
"5 editing tips" is generic. "5 editing mistakes I made in my first 50 videos" can only come from you.
The same idea, pushed
Generic: "Post about meal prepping." Pushed: "The 3 meal-prep shortcuts I actually kept after a year of trying everything — for parents with 20 minutes on a Sunday." Same topic. One of them sounds like a person.
Pause and think: Think of your best-performing piece ever. Which of the three moves — constraint, angle, experience — was quietly doing the work?
Checkpoint
Prompt exercise
Generate twenty ideas that actually fit your niche
Copy this prompt into ChatGPT, Claude, Gemini, Copilot, or whichever AI tool you have access to — the website doesn't run AI itself. Answer its questions honestly; the specifics are what make the output yours.
Act as a content strategist who helps creators find ideas that fit their specific niche. Ask me one question at a time: first my niche (push me to narrow it if it's broad), then who my audience is and what they struggle with, then my 3 best-performing pieces and why I think they worked. After that, generate 20 content ideas grouped into four types: how-tos, personal stories, opinions, and behind-the-scenes. For each idea, add one line on why MY audience specifically would care. No generic ideas that any creator could post.
Reflection: Out of twenty, how many passed the "would my audience actually care?" filter — and did the keepers share a pattern? That pattern is worth writing down.
Quick check
3 quick questions — no pressure
There's no pass or fail here. Answer them all, and we'll show you the answers either way.