- The Vibe Marketer
- Posts
- Building your audience from scratch doesn't have to be hard (a deep dive)
Building your audience from scratch doesn't have to be hard (a deep dive)
4-stage system: gather, find your gap, build your pillars, post
Hey Vibe Marketer,
“Building your audience from scratch for your niche does not have to be hard.”
I saw a common question that kept coming up inside the vibe marketer’s community: "How do I build my audience on LinkedIn or X from scratch?"
That got me thinking about it.
Let me break your first inhibition. You don't have to reinvent the wheel. There are already creators sitting at 100K+ in your niche their entire playbook is public, and you can just learn from it.
Here’s how you can approach it in stages:
#1 Lock your pre-requisites: your niche, your person, your channel, your voice baseline. Before you touch a single scraper.
#2 Gather and analyze: identify the right creators for your stage, pull their data, and go all the way down to their copy structure: hooks, sentence patterns, CTAs, not just topics.
#3 Find your arbitrage: the positioning gap, the timing window, the content format nobody in your niche is using yet.
#4 Build your first 30 days: turn the arbitrage into content pillars, then a calendar, then real posts that generate your own data.
That's it. That's the whole system.
In today's issue, we break down each part of this system step by step, with the exact questions to ask at every stage.
Take this exact framework, hand it to AI, and you can have this running as a working skill in under 30 minutes. Here's the skill creator skill to get started.
Stage 0: Pre-requisites (Starting From Scratch)
Before you get started with your research, lock the following:
a. Your niche: the specific topic, not a broad category.
b. Your audience: the exact reader you're writing for. Not "marketers." A marketer drowning in tool fatigue at a 10-person agency.
c. Your channel: pick one to start, not several at once.
d. Your voice baseline: even rough. You need something to filter borrowed patterns through, or you'll just clone someone else's account.
e. Your proof: whatever you already have to back up what you'll be talking about. A website, past work, a portfolio, screenshots of results. Have it ready before you need to point to it.
Stage 1: Gather and Analyze
1. Identify your creators : by tier, matched to your own stage.
You can't take inspiration from someone at 100K when you're starting from zero. Their content, their authority, their audience trust none of it transfers cleanly to your starting point. Bucket the creators in your niche by follower count, and match the bucket to where you actually are:
→ Starting from scratch → study creators in the 3,000-7,000 range
→ Scaling past your first few thousand - study creators one or two tiers up from where you are
→ Already established - study the 50K-100K+ tier
Then filter that bucket further. Followers alone don't qualify a creator for your research set check how active they actually are. When did they last post? What's their posting cadence looked like recently? An account with 8,000 followers that hasn't posted in four months isn't a useful data point.
2. Gather the data on that filtered list. For each creator, pull their top-performing posts and their positioning how they describe themselves, who they're clearly talking to, what they're known for in the niche. | 3. Analyze what you've gathered. Narrow from the full set down to specifics: what topics come up again and again, what formats and media types they lean on, what CTA patterns repeat. You're going from a broad bucket of creators to a specific, narrow answer. This is what a winning format looks like, this is what a winning content pillar looks like, this is how their messaging ties back to their positioning. |
Don't stop at topic and format go down to the copy structure itself. This is the layer most people skip, and it's where the actual mechanics live:
Check | What to look for | What it tells you |
|---|---|---|
Hooks | Recognizable structure PAS (Problem, Agitate, Solution), AIDA (Attention, Interest, Desire, Action), contrarian-claim-then-proof, or something else | Most winning hooks in a niche cluster around 2-3 repeatable structures, not an infinite variety |
Sentence and line length | Short, fragmented lines versus longer flowing sentences | Varies by niche and by pillar type measurable across your research set, not a guess |
CTA structure | Soft ("let me know what you think") versus direct ("try this and reply with your result") versus none at all | Track which shows up on the posts with real engagement, not just any post |
Tools:
Need | Use | Note |
|---|---|---|
Pull the raw dataset | A scraping actor for your channel. Apify has purpose-built actors depending on the channel you're targeting | Don't force tools you don't need. The point is the dataset, not the tool stack |
Process what you pull | Claude Code or another AI tool | This is where decomposition and copy-structure tagging actually happens |
Broader web context, if needed | Firecrawl or an MCP | Optional, only if the scraper alone isn't enough |
That's Stage 1 in full: identify creators at the right tier, filter by activity, gather their data, analyze it down to specifics. What you do with those specifics comes next.
Stage 2: Find Your Arbitrage
Gathering and analyzing gets you a map of what's already working. This stage is where you find the gap in that map the opening nobody in your research set is filling yet.
Arbitrage | The question to ask | What to do with the answer |
|---|---|---|
1. Timing | Is this topic still in the window where only a handful of people are covering it, before it peaks? | Whatever surfaces trends for you, run it regularly. Treat "almost nobody's covering this yet" as the strongest signal catching the window is a skill, not luck |
2. Positioning | How do the creators in your research set position themselves? What's the common ground, and where's the gap? | If everyone in your niche and tier positions the same way, that's not a lane to copy. It's a lane to differentiate from. Your arbitrage is the positioning nobody's claimed yet |
3. Content theme | If the creators you researched all lean on one format, where's the scope to mix and match what they're not doing? | Combine it with your positioning arbitrage from step 2. Your positioning gap, applied through a format gap, is what makes your content different instead of a copy |
4. Copy structure library | Which hook structure, which format, and which media type actually perform together in your research set. Not in isolation? | Build a simple reference (an MD file library in Claude Code works well), organized by the copy structures you found repeating. Match combinations against your content pillars, so you're pulling from a tested combination instead of guessing at structure every time |
A worked example for #4: a contrarian hook might pair with short text and no image in your research set, while a PAS structure might pair with carousel or long-form. That's the kind of combination the library should surface.
This scrape-and-match approach, including the idea that what resonates on one channel can be repurposed for another, is the same play broken down in this playbook.
Stage 3: Your First 30 Days
Once you know what's actually working in your niche and where your arbitrage is, three things happen before you post not after.
1. Build your content pillars.
A pillar isn't a topic. "[your niche]" is a topic — too broad to post against every week without repeating yourself or drifting off-niche. Every pillar you build runs on the same formula:
Pillar = your niche + your positioning arbitrage + the wave you caught early + the market gap you found.
Run that formula four times, with a different lead input each time, and four distinct pillars fall out:
Content pillar framework
2. Equip yourself. Before you fill a calendar, you need to be able to speak with something real behind it. This means staying current on your wave you can't credibly cover a trend you stopped tracking two weeks ago and positioning yourself as someone who's a chapter ahead, not the world's foremost expert. This isn't a one-time prep step. It runs alongside everything else, continuously.
3. Build your calendar. Fill it by rotating through the four pillar
Week | Pillar focus | Cadence | What you're testing |
|---|---|---|---|
Week 1 | Post against all 4 pillars, using your Stage 2 hypothesis unchanged | 4-5 posts | Does the borrowed pattern get any traction at all in your voice, across each pillar? |
Week 2 | Keep the same 4 pillars, vary the hook only within each | 4-5 posts | Which hook style your specific audience responds to, per pillar |
Week 3 | Keep hook style constant, vary format/length per pillar | 4-5 posts | Whether structure or length is the bigger lever, and whether that differs by pillar |
Week 4 | Start tagging your own posts by component, pillar by pillar (opener → impressions, body → likes, CTA → comment %) | 4-5 posts | You now have ~15-20 posts of your own data, split across 4 pillars — enough to see which pillar is actually pulling weight |
By day 30, you switch off borrowed patterns. From here, you're running the Justin Welsh method on your own results, per pillar:
1. track opener performance by impressions
2. body content by likes (read-through proxy)
3. CTA by comment percentage
4. Re-test quarterly: the pattern decays as the algorithm and your audience shift
5. And Timing decays fastest of the four.
It's a hypothesis with an expiration date, not a permanent template.
Wrapping Up
That's the full system: lock your pre-requisites, gather and analyze the creators already proving what works, find your arbitrage in the gaps they've left, then build your own 30 days on top of it.
One honest note before you run with it. Stage 1 or Stage 2 is a starting hypothesis, built from studying what's already worked for others.
The part that actually compounds is Stage 3: posting against your own pillars, long enough to generate your own data, and letting that data tell you what to do next.
You don't need to reinvent the wheel to start.
If you're trying to build your audience or grow your channel or looking for ways to work with AI for marketing, my next live session with the vibe marketers inside the community is tomorrow.
I do this bi-weekly, where community members discuss, exchange notes, showcase builds, and come on hot seats and get their real business problems solved.
If you've been building with AI alone and you're looking for AI marketing practitioners who are figuring this out in real time, solving real problems, shipping real things, this is your place.
—
Miscellaneous vibes
1. Turn your call with prospect or customer into leverage, content and marketing asset…
2. Boring marketer is opening 3 ai marketing advisory slots.
3. Agents are about to outnumber humans on the internet?
4. 6-step AI playbook for local SEO that prints leads…
5. Why are some AI companies hitting $10M ARR* in six months?
6. What is a brand context loop?
7. 7 use cases to turn hermes agent into a full-stack marketing team…
Let’s vibe
Crazy time to be building with AI.
Wanna know if your business is showing up on ChatGPT? Run your audit.
Wanna scale marketing and outperform your competitors? Upgrade your AI.
Wanna learn from marketers actually shipping with AI? Upgrade yourself.
Wanna unlock both at once? Get the bundle.
Got feedback or a question? Just hit reply. I read every message.
Know someone who’d love this newsletter? Share it with a friend who is figuring AI out too.
The Vibe Marketer
How'd you like this letter?Was this a vibe? |