[Journey] From $0 to… Hopefully Something – My OFM + AI Experiment

[SS2211]

Great progress so far. You seem to be constantly improving and always learning from what goes well and what doesn't so I'm sure you will do really well.

It's interesting to see how you now go from grinding yourself to start managing everything once your system falls in place. I've been following a business doing something completely different than OF for a while. The always talk how important about their system is. They try to reduce any friction they can. Even the smallest thing is worth analyzing and improving on so things go smoothly. It's so important in any business to make things flow. Once a business have a good system in place it can easily add new people and keep on growing.

 

[shijin]

Great progress so far. You seem to be constantly improving and always learning from what goes well and what doesn't so I'm sure you will do really well.

It's interesting to see how you now go from grinding yourself to start managing everything once your system falls in place. I've been following a business doing something completely different than OF for a while. The always talk how important about their system is. They try to reduce any friction they can. Even the smallest thing is worth analyzing and improving on so things go smoothly. It's so important in any business to make things flow. Once a business have a good system in place it can easily add new people and keep on growing.

You’re absolutely right about the system part. In the beginning, I was doing everything manually and reacting to every little issue like it was a fire. Now, the more I refine the workflow, from content scheduling to AI chat templates. the smoother everything gets. It’s wild how even small tweaks (like automating tiny repetitive steps or cleaning up tracking sheets) can free up hours and mental energy. Once the “machine” starts running itself a bit, it’s way easier to think big picture instead of just surviving day to day.
Totally agree, any business, not just OFM, lives or dies by how well the system runs behind the scenes.

 

[shijin]

Day 41: The Calm After the Chaos ️

If last week felt like sprinting through a storm, today finally felt… balanced.
No huge spikes, no platform drama — just steady progress and a sense that things are *finally* clicking together.

Revenue
Pulled in $462 today — not the highest number this week, but it felt good because it came without stress. No firefighting, no late-night panic fixes. Just the system quietly doing its thing.

Recurring subs are starting to make up around 40% of the daily total now, which is such a relief. That passive base makes everything easier.

‍ Models
Model #1: Took the day off filming, but her pinned TikTok is still performing strong (25k views in 24 hours).
Model #2: Posted her first “interactive” story poll — engagement jumped instantly.
Model #3: Landed a $140 tip from a returning fan who said he “missed her energy.” That one message made my whole morning.
Model #4: Still finding her footing but getting more confident each day.

I’m realizing it’s not about adding more models fast — it’s about making sure the ones I have are truly optimized. Quality > quantity is starting to feel real.

AI & Chat Flow
I added a new “tone shifter” option today. Basically, AI can switch between friendly / flirty / VIP-exclusive tone depending on the fan’s tag.
It sounds small, but it’s been a game changer for keeping convos fresh.

Also trained the chatter on a few “story arcs” — like weekend plans, random life updates — so the convos feel human instead of scripted. The feedback loop between AI and human chatter is starting to feel seamless.

Traffic & Ads
TikTok still steady — no viral spike, but engagement’s consistent.
Meta ads are running clean again after I toned down visuals and focused on storytelling captions.
Reddit’s heating up again — I experimented with reaction-style captions and the CTR jumped 22%.

Planning to test Pinterest next week. I don’t expect miracles, but who knows — the algorithm gods work in mysterious ways.

Reflections
The biggest shift lately isn’t technical — it’s mental. I’ve gone from *doing everything myself* to *trusting the process* and delegating.
It’s weirdly uncomfortable but freeing.

Every time I hand off a task (posting, chatting, tracking), my first instinct is “they won’t do it right.”
But they almost always do — sometimes even better.
That realization alone is making this whole thing start to feel like an actual business rather than a solo hustle.

Tomorrow
- Refresh ad creatives with new visuals from Model #3
- Schedule TikTok posts for the weekend
- Finish drafting the 6-week reflection post
- Take a couple hours to actually rest — my eyes are begging for it

Forty-one days in and it’s starting to feel like I’m finally steering the ship instead of paddling behind it.

 

[shijin]

Day 42: A Bump in the Road... and a Quick Recovery

You ever have one of those days where everything *technically works*, but somehow it all feels a little off?
That was today. Not a disaster — just one of those weird mid-week wobbles that remind you this whole thing’s still very human.

Daily Revenue
Ended the day at $398 — a dip from the past few days, but totally expected after a hot streak.
A few subs expired without renewals, and one big tipper went silent (hoping he circles back soon).

Still, even on an “off” day, it’s wild that numbers like this now feel normal. Perspective is everything.

‍ Models
Model #1: Re-shot a reel that got flagged for “suggestive content.” The new version actually looks way more natural — happy accident.
Model #2: Experimented with a new post timing window (1–3PM EST) and saw a small engagement bump — might stick with that.
Model #3: Her fans are super loyal. She got 2 spontaneous tips today after replying to some older messages.
Model #4: Had some content delays — her lighting setup broke, so we had to skip a post. But she’s handling it like a pro.

Traffic & Ads
TikTok slowed down slightly — looks like the algo rotated again. One reel that I *thought* would do well got under 3k views, which stung a bit.
Meta ads are steady — $25 spend brought in 6 leads, one of which turned into a serious chat.
Reddit continues to surprise me — I’m getting better at reading which subreddits tolerate certain post formats. “Soft storytelling” captions seem to perform best right now.

AI & Chat Flow
Tweaked a few prompts today — I noticed the AI was repeating certain phrasing (“you’re so sweet” kept showing up too often ).
Added a tiny randomization layer that swaps out similar responses so chats feel more spontaneous. Early results look great — fans are chatting longer again.

Also started testing a “mood tracker” for fans — tagging convos as warm/cold/interested — to prioritize replies better. Already saving time.

Reflections
If Day 40 felt like control, Day 42 was a reminder that even the best system needs human eyes.
Small dips used to stress me out a ton, but now they just feel like data.
The difference between a good day and a bad day isn’t the numbers — it’s how you react to them.

Tomorrow’s Plan
- Refresh ad creatives (new captions, less text-heavy)
- Schedule content drops for the weekend
- Have a quick call with Model #4 about her new theme idea
- Take an hour to clean up the chat tracking sheet — it’s starting to look like spaghetti

Not every day is a win on paper, but every day still moves the whole thing forward.
And that’s starting to feel like the real flex.

 

[tomschneider55]

Day 42: A Bump in the Road... and a Quick Recovery

You ever have one of those days where everything *technically works*, but somehow it all feels a little off?
That was today. Not a disaster — just one of those weird mid-week wobbles that remind you this whole thing’s still very human.

Daily Revenue
Ended the day at $398 — a dip from the past few days, but totally expected after a hot streak.
A few subs expired without renewals, and one big tipper went silent (hoping he circles back soon).

Still, even on an “off” day, it’s wild that numbers like this now feel normal. Perspective is everything.

‍ Models
Model #1: Re-shot a reel that got flagged for “suggestive content.” The new version actually looks way more natural — happy accident.
Model #2: Experimented with a new post timing window (1–3PM EST) and saw a small engagement bump — might stick with that.
Model #3: Her fans are super loyal. She got 2 spontaneous tips today after replying to some older messages.
Model #4: Had some content delays — her lighting setup broke, so we had to skip a post. But she’s handling it like a pro.

Traffic & Ads
TikTok slowed down slightly — looks like the algo rotated again. One reel that I *thought* would do well got under 3k views, which stung a bit.
Meta ads are steady — $25 spend brought in 6 leads, one of which turned into a serious chat.
Reddit continues to surprise me — I’m getting better at reading which subreddits tolerate certain post formats. “Soft storytelling” captions seem to perform best right now.

AI & Chat Flow
Tweaked a few prompts today — I noticed the AI was repeating certain phrasing (“you’re so sweet” kept showing up too often ).
Added a tiny randomization layer that swaps out similar responses so chats feel more spontaneous. Early results look great — fans are chatting longer again.

Also started testing a “mood tracker” for fans — tagging convos as warm/cold/interested — to prioritize replies better. Already saving time.

Reflections
If Day 40 felt like control, Day 42 was a reminder that even the best system needs human eyes.
Small dips used to stress me out a ton, but now they just feel like data.
The difference between a good day and a bad day isn’t the numbers — it’s how you react to them.

Tomorrow’s Plan
- Refresh ad creatives (new captions, less text-heavy)
- Schedule content drops for the weekend
- Have a quick call with Model #4 about her new theme idea
- Take an hour to clean up the chat tracking sheet — it’s starting to look like spaghetti

Not every day is a win on paper, but every day still moves the whole thing forward.
And that’s starting to feel like the real flex.

Hallo Shijin,

love your journey and your progress. Keep building. It looks like you are onto something here!

My questions:

1. Who do you think are your main competitors - Are you not scared that some of them will text your models and snap them away?
2. What kind of software are the big players using? Also something they build themselves or some other infrastructure?
3. Do you use n8n or which funnel tool?
4. Which LLM are you for nsfw?
5. The meta ads are leading to where? Straight to OF or to a a landing page?

 

[shijin]

Hallo Shijin,

love your journey and your progress. Keep building. It looks like you are onto something here!

My questions:

1. Who do you think are your main competitors - Are you not scared that some of them will text your models and snap them away?
2. What kind of software are the big players using? Also something they build themselves or some other infrastructure?
3. Do you use n8n or which funnel tool?
4. Which LLM are you for nsfw?
5. The meta ads are leading to where? Straight to OF or to a a landing page?

Appreciate it a lot, man, really glad you’re following along!
To your questions:

1. Honestly, competition’s always there. The space is packed with small agencies and solo managers now. But I’m not too worried about people trying to poach models, I build real relationships, share revenue transparently, and make sure they see consistent results. When they feel supported and make money, they’re not looking to leave.
2. Most of the “big players” use custom-built dashboards or layered systems that tie together analytics, chat tools, and content schedulers. A lot of them started on Zapier or n8n, then upgraded once revenue justified it. I’m halfway there, using a lightweight version built around automation and GPT helpers, but planning something more custom later.
3. Yep, I use n8n, it’s the brain behind my small automations. It handles posting schedules, alert triggers, and a few DM follow-ups. Way cheaper and more flexible than Zapier.
4. For AI, I use a mix of LLMs (nothing NSFW-tuned directly). I build prompts that keep the tone suggestive but not explicit, so it stays compliant and human-sounding. The trick isn’t the model, it’s how you prompt and personalize it.
5. Meta ads go to a landing page, never directly to OF. The landing page acts as a “creator application” or funnel, filtering leads and giving me a chance to qualify before onboarding. It’s safer and converts better than sending traffic straight to OF.

So far, that combo, organic + n8n automations + warm landing funnel, has been the sweet spot.

Respect for actually starting instead of just reading. Most people never even make it to this first post. Keep documenting everything even the flops. That’s where the gold is.

You’re totally right. The best lessons always come from the messy parts, not the highlight moments. I’ve been trying to share the wins and the screw-ups because honestly, that’s what I wish I had when I first started reading other journeys.
The “flops” end up teaching you more than anything, and sometimes those rough days are what lead to the biggest breakthroughs later.

 

[azbotboyz]

Niche journey, what device/browser you use for managing account and posting ?

 

[tomschneider55]

Appreciate it a lot, man, really glad you’re following along!
To your questions:

1. Honestly, competition’s always there. The space is packed with small agencies and solo managers now. But I’m not too worried about people trying to poach models, I build real relationships, share revenue transparently, and make sure they see consistent results. When they feel supported and make money, they’re not looking to leave.
2. Most of the “big players” use custom-built dashboards or layered systems that tie together analytics, chat tools, and content schedulers. A lot of them started on Zapier or n8n, then upgraded once revenue justified it. I’m halfway there, using a lightweight version built around automation and GPT helpers, but planning something more custom later.
3. Yep, I use n8n, it’s the brain behind my small automations. It handles posting schedules, alert triggers, and a few DM follow-ups. Way cheaper and more flexible than Zapier.
4. For AI, I use a mix of LLMs (nothing NSFW-tuned directly). I build prompts that keep the tone suggestive but not explicit, so it stays compliant and human-sounding. The trick isn’t the model, it’s how you prompt and personalize it.
5. Meta ads go to a landing page, never directly to OF. The landing page acts as a “creator application” or funnel, filtering leads and giving me a chance to qualify before onboarding. It’s safer and converts better than sending traffic straight to OF.

So far, that combo, organic + n8n automations + warm landing funnel, has been the sweet spot.

You’re totally right. The best lessons always come from the messy parts, not the highlight moments. I’ve been trying to share the wins and the screw-ups because honestly, that’s what I wish I had when I first started reading other journeys.
The “flops” end up teaching you more than anything, and sometimes those rough days are what lead to the biggest breakthroughs later.

Thanks for that quick response! Really appreciate your honesty.

So whats the goal - i know you just wanted to get a feel and try it out. I am sure you thinking of a long term goal or numbers. You think revenues of a 100k a month is possible or to much upscaling?

 

[Henrykbrodaty]

Hey bro, I’ve got couple of questions. What nationality are your models? And what do you think about marketplaces where you can find models?do your girls have access to the accounts?

 

[shijin]

Day 43: A Little Momentum Goes a Long Way

After yesterday’s slowdown, I woke up ready to shake things up — and the universe actually delivered.
No huge miracle moment, but a few small wins stacked up in the best way possible.

Daily Revenue
Closed the day at $512
That’s a nice bounce from yesterday’s dip. A few big tips came through, and a couple of renewals hit right before midnight, which pushed it over the $500 mark again.

The coolest part? Almost all of it was from *returning fans*. Retention’s officially becoming the backbone of this whole system.

‍ Models
Model #1: Her new reel finally picked up traction after two quiet days — 42k views in under 24 hours. Engagement’s back, and you can feel her confidence returning.
Model #2: Her “behind-the-scenes” clips are working wonders. She’s getting new followers who actually comment instead of just scrolling.
Model #3: One of her older TikToks randomly got picked up again by the algorithm — went from 3k views to 60k overnight. That one post brought in three new subs directly.
Model #4: Shot her first mini content series (5 short clips tied together). She’s a bit nervous, but I can see potential — her tone and camera presence are improving fast.

It’s starting to feel less like managing four separate people and more like running a creative studio.

Traffic & Ads
Meta ads performed surprisingly well today — $25 spend → 8 new leads → 3 good convos.
Still not scaling aggressively, but the ROI is looking better every day.
I also tested a new funnel variation with a cleaner layout and less text — early stats show a lower bounce rate already.

TikTok’s holding steady, and Reddit’s still a goldmine when I time posts right.
Fun experiment: I tried adding subtle “story captions” (like “this moment changed her day”) and engagement almost doubled. People love narrative even in short clips.

AI & Chat Flow
I gave the AI a bit more creative freedom today — loosened the tone rules slightly so it sounds more spontaneous. The result? Conversations feel way more “alive.”
Fans are staying longer in DMs and replying faster.

Also started a tiny new experiment: “AI memory moments.”
Basically, the system drops in small references to past chats (like “you said you were working late yesterday, how’d that go?”). That one tweak makes people think it’s 100% human.

Reflection
Today reminded me that growth isn’t linear — some days you grind for small wins, and other days everything quietly clicks back into place.
I’m starting to appreciate the steady middle ground more than the rollercoaster.

The system is stabilizing. The team feels confident.
And even though the journey’s still messy, I can actually see the shape of what it’s becoming.

Tomorrow
- Refresh TikTok captions and try 2 new hooks
- Launch the new ad set for model outreach
- Review fan retention data and tweak chat follow-ups
- Maybe, finally, get a full night of sleep

It’s funny — the best days aren’t the explosive ones anymore.
It’s the ones where everything quietly *works*.
That’s when you know you’re building something real.

 

[shijin]

Niche journey, what device/browser you use for managing account and posting ?

Good question, I try to keep everything separated and clean to avoid any platform issues.
For most account management, I use Shift + multiple Chrome profiles with separate cookies and extensions, each model gets her own environment. For posting, I stick to a dedicated phone per 1–2 models so the activity looks organic and consistent (same IPs, same behavior patterns).
It’s a bit of a hassle at first, but it massively reduces the risk of shadowbans or logouts. Stability > convenience in this game.

Thanks for that quick response! Really appreciate your honesty.

So whats the goal - i know you just wanted to get a feel and try it out. I am sure you thinking of a long term goal or numbers. You think revenues of a 100k a month is possible or to much upscaling?

Yeah, when I first started, I honestly just wanted to see if the system would even work. But now that it’s running smoothly and bringing in consistent revenue, I’m definitely thinking more long term.
$100K/month is ambitious, but not impossible, it just requires solid infrastructure, more human help, and tighter systems. The bottleneck isn’t demand; it’s how fast you can scale without breaking quality.
Right now, my short-term goal is to stabilize at $20K/month net while refining the backend (ads, chat flow, retention). Once that’s smooth and repeatable, scaling to six figures becomes more of a math problem than a dream. Step by step.

Hey bro, I’ve got couple of questions. What nationality are your models? And what do you think about marketplaces where you can find models?do your girls have access to the accounts?

Most of the models I work with are from Europe and South America, mainly because communication is smooth, and they tend to be super reliable with content delivery. I’ve found it’s less about nationality and more about consistency and mindset. Marketplaces can work, but they’re pricey, a lot of them take big commissions or have upfront fees that don’t make sense when you’re still scaling. I prefer direct outreach since it’s cheaper and lets me build trust from the start.
As for access, yeah, the girls have login access to their accounts, but I handle most of the management (posting, messaging, analytics, etc.). Transparency’s key. That way they still feel ownership, but I can keep everything running efficiently.

 

[shijin]

Day 44: When an Ad Finally Clicks

I swear, after weeks of tweaking captions, swapping thumbnails, and overthinking targeting — one of the Meta ads finally decided to wake up.
It’s been quietly outperforming every test I’ve run so far, and it feels *so good* to finally see the numbers reflect the effort.

Daily Revenue
Closed the day at $582.
Biggest push came from new subscribers funneling in through the improved ad → landing page → DM sequence.
What’s crazy is that this was *only* a $20 test ad. The conversion rate more than doubled overnight.

The Ad That Worked
After testing a dozen angles, the one that hit wasn’t flashy at all — just a simple “creator story” format:
> “She started with nothing but a camera and a dream. 30 days later, she’s running her own brand.”

No spicy photos, no clickbait — just authentic storytelling. It pulled a 4.6% CTR, which is way higher than the usual 2–3%.
The landing page follow-up messages are converting better too. Sometimes less is more.

‍ Models
Model #1: Back in her groove. Posted a short “Day in the Life” vlog, which did great — 55k views, lots of genuine comments.
Model #2: She’s starting to bring in her own content ideas now, which is awesome. Makes everything feel more natural.
Model #3: Still the top performer — she’s got that magic mix of confidence + authenticity that fans can’t resist.
Model #4: Did her first themed set today and honestly crushed it. You can tell she’s finally comfortable on camera.

It’s funny watching how each model’s personality attracts a different type of fan. Feels like running four mini-brands under one roof.

AI & System Updates
The AI’s new “conversation memory” is paying off big time. Fans are actually referencing past chats like they’re talking to the same person every day.
That personal touch keeps them coming back — and tipping.

Also, I automated a “Good Morning” and “Good Night” system for top fans. Super basic, but engagement jumped instantly. It’s wild how something so simple works so well.

Traffic
TikTok’s been decent — a few reels hit 30k+, and engagement’s holding steady.
Reddit… continues to be Reddit. A few posts flopped, one post blew up — can’t predict it, just gotta stay consistent.
Pinterest experiment is in early stages, but the impressions are coming faster than expected. Might actually have potential there.

Reflections
This ad test win feels like a turning point. For the first time, I can see a clear, repeatable path from cold traffic → lead → model onboarding → fan engagement → revenue.
It’s not perfect, but it’s finally scalable.

Every time I fix one small bottleneck — whether it’s ad performance, chat flow, or content scheduling — everything downstream gets smoother.
That’s the real game: stacking tiny improvements until the system starts to breathe on its own.

Tomorrow
- Scale the winning ad slowly ($20 → $50) to see if it holds
- Set up the next version of the landing page with updated copy
- Help Model #4 shoot her second content set
- Prep early ideas for a “Month Two” highlight recap post

Sometimes all it takes is one ad working to remind you that the grind *does* pay off.
Feeling a little extra motivated tonight.

 

[shijin]

Day 44: When an Ad Finally Clicks

I swear, after weeks of tweaking captions, swapping thumbnails, and overthinking targeting — one of the Meta ads finally decided to wake up.
It’s been quietly outperforming every test I’ve run so far, and it feels *so good* to finally see the numbers reflect the effort.

Daily Revenue
Closed the day at $582.
Biggest push came from new subscribers funneling in through the improved ad → landing page → DM sequence.
What’s crazy is that this was *only* a $20 test ad. The conversion rate more than doubled overnight.

The Ad That Worked
After testing a dozen angles, the one that hit wasn’t flashy at all — just a simple “creator story” format:
> “She started with nothing but a camera and a dream. 30 days later, she’s running her own brand.”

No spicy photos, no clickbait — just authentic storytelling. It pulled a 4.6% CTR, which is way higher than the usual 2–3%.
The landing page follow-up messages are converting better too. Sometimes less is more.

‍ Models
Model #1: Back in her groove. Posted a short “Day in the Life” vlog, which did great — 55k views, lots of genuine comments.
Model #2: She’s starting to bring in her own content ideas now, which is awesome. Makes everything feel more natural.
Model #3: Still the top performer — she’s got that magic mix of confidence + authenticity that fans can’t resist.
Model #4: Did her first themed set today and honestly crushed it. You can tell she’s finally comfortable on camera.

It’s funny watching how each model’s personality attracts a different type of fan. Feels like running four mini-brands under one roof.

AI & System Updates
The AI’s new “conversation memory” is paying off big time. Fans are actually referencing past chats like they’re talking to the same person every day.
That personal touch keeps them coming back — and tipping.

Also, I automated a “Good Morning” and “Good Night” system for top fans. Super basic, but engagement jumped instantly. It’s wild how something so simple works so well.

Traffic
TikTok’s been decent — a few reels hit 30k+, and engagement’s holding steady.
Reddit… continues to be Reddit. A few posts flopped, one post blew up — can’t predict it, just gotta stay consistent.
Pinterest experiment is in early stages, but the impressions are coming faster than expected. Might actually have potential there.

Reflections
This ad test win feels like a turning point. For the first time, I can see a clear, repeatable path from cold traffic → lead → model onboarding → fan engagement → revenue.
It’s not perfect, but it’s finally scalable.

Every time I fix one small bottleneck — whether it’s ad performance, chat flow, or content scheduling — everything downstream gets smoother.
That’s the real game: stacking tiny improvements until the system starts to breathe on its own.

Tomorrow
- Scale the winning ad slowly ($20 → $50) to see if it holds
- Set up the next version of the landing page with updated copy
- Help Model #4 shoot her second content set
- Prep early ideas for a “Month Two” highlight recap post

Sometimes all it takes is one ad working to remind you that the grind *does* pay off.
Feeling a little extra motivated tonight.

 

[shijin]

Day 45: Halfway to Something Real

Forty-five days in.
It’s weird — at first, I was counting days like milestones… now they just blend together into this rhythm.
There’s still chaos, sure, but it’s a *controlled* kind of chaos now.

Daily Revenue
Wrapped up the day at $547.
Not a new high, but what made it special was how *stable* it felt.
No flukes, no sudden drops — just steady, predictable income from all the pieces working together.

The System in Motion
That winning Meta ad from yesterday is still going strong — CTR’s holding at 4.2%, leads cost around $0.10–$0.12 each, and 3 new quality chats came through today.
It’s wild how much difference one solid funnel can make once it clicks.

TikTok’s doing its thing too — nothing viral, but 2–3 clips consistently hit 20–30k views daily, and traffic stays healthy.
Reddit’s surprisingly calm this week, so I’m using that breathing room to optimize content scheduling instead of spamming.

‍ Models
Model #1: Her confidence is back — she filmed a new series of lifestyle clips that feel *so natural*. I’m betting they’ll perform better than anything super polished.
Model #2: She’s officially the “organized one” — sends her entire week’s content in one go, makes my life a hundred times easier.
Model #3: Still the top earner, no surprise there. Her regulars are loyal and keep the chats flowing.
Model #4: Getting there — her audience is smaller, but the conversion rate’s higher. Sometimes smaller + loyal beats big + flaky.

I’m starting to realize each model is basically their own little ecosystem — you just have to learn what keeps it healthy.

AI & Workflow
Refined the chat prompts again — added light personality “modes” that shift tone based on how long the fan’s been subscribed.
Day 1 fan? Friendly and flirty.
Week 3 fan? More emotional, familiar tone.
Month 2 fan? VIP treatment.
The engagement difference is clear. Feels like I’m finally tapping into relationship-style retention rather than one-off chats.

Also tested a new tool to sync DM templates and analytics between chatter and AI. It’s not perfect yet, but it saves time switching windows.

Reflections
If the first month was about *figuring it out*, this second one’s about *making it real*.
I’m spending less time experimenting and more time refining — fixing small cracks before they turn into big ones.

The wild part? It’s starting to feel sustainable. Like something I could actually scale long term without losing my mind.
Still a lot to learn, still a ton of things to automate better, but the foundation feels strong now.

Tomorrow
- Scale the ad budget again to $60 and watch for cost jumps
- Start onboarding prep for Model #5 (if I can keep up )
- Brainstorm TikTok “duet” content for Model #3
- Write the Month 2 Recap draft — want it to hit hard

Forty-five days in, and I can finally see what this could become if I just keep showing up.
Not there yet — but definitely not where I started.

 

[PINKAI]

Good Luck for your journey!

 

[PINKAI]

Day 45: Halfway to Something Real

Forty-five days in.
It’s weird — at first, I was counting days like milestones… now they just blend together into this rhythm.
There’s still chaos, sure, but it’s a *controlled* kind of chaos now.

Daily Revenue
Wrapped up the day at $547.
Not a new high, but what made it special was how *stable* it felt.
No flukes, no sudden drops — just steady, predictable income from all the pieces working together.

The System in Motion
That winning Meta ad from yesterday is still going strong — CTR’s holding at 4.2%, leads cost around $0.10–$0.12 each, and 3 new quality chats came through today.
It’s wild how much difference one solid funnel can make once it clicks.

TikTok’s doing its thing too — nothing viral, but 2–3 clips consistently hit 20–30k views daily, and traffic stays healthy.
Reddit’s surprisingly calm this week, so I’m using that breathing room to optimize content scheduling instead of spamming.

‍ Models
Model #1: Her confidence is back — she filmed a new series of lifestyle clips that feel *so natural*. I’m betting they’ll perform better than anything super polished.
Model #2: She’s officially the “organized one” — sends her entire week’s content in one go, makes my life a hundred times easier.
Model #3: Still the top earner, no surprise there. Her regulars are loyal and keep the chats flowing.
Model #4: Getting there — her audience is smaller, but the conversion rate’s higher. Sometimes smaller + loyal beats big + flaky.

I’m starting to realize each model is basically their own little ecosystem — you just have to learn what keeps it healthy.

AI & Workflow
Refined the chat prompts again — added light personality “modes” that shift tone based on how long the fan’s been subscribed.
Day 1 fan? Friendly and flirty.
Week 3 fan? More emotional, familiar tone.
Month 2 fan? VIP treatment.
The engagement difference is clear. Feels like I’m finally tapping into relationship-style retention rather than one-off chats.

Also tested a new tool to sync DM templates and analytics between chatter and AI. It’s not perfect yet, but it saves time switching windows.

Reflections
If the first month was about *figuring it out*, this second one’s about *making it real*.
I’m spending less time experimenting and more time refining — fixing small cracks before they turn into big ones.

The wild part? It’s starting to feel sustainable. Like something I could actually scale long term without losing my mind.
Still a lot to learn, still a ton of things to automate better, but the foundation feels strong now.

Tomorrow
- Scale the ad budget again to $60 and watch for cost jumps
- Start onboarding prep for Model #5 (if I can keep up )
- Brainstorm TikTok “duet” content for Model #3
- Write the Month 2 Recap draft — want it to hit hard

Forty-five days in, and I can finally see what this could become if I just keep showing up.
Not there yet — but definitely not where I started.

are you using AI for chatting as well? or engaging with followers via DM? or simply link in Bio or auto reply?

 

[MarcoKayzimir]

Amazing. I've never thought about AI from such the angle. Thank you for sharing and good luck!

 

[VerifySetLLC]

Good luck, i follow your jouerney.

 

[shijin]

are you using AI for chatting as well? or engaging with followers via DM? or simply link in Bio or auto reply?

Yeah, I use AI for chatting too, but it’s more of a co-pilot than a full autopilot. The AI drafts replies based on the fan’s tone and chat history, then I (or my chatter) skim and tweak before sending. That way it still feels human and personal.
For engagement, it’s a mix: some followers come through DMs, others via the link in bio. I also have light auto-replies set up for new followers just to start the convo, but everything meaningful still goes through human eyes. It’s that blend of AI efficiency + human touch that keeps it natural and avoids getting flagged.

Amazing. I've never thought about AI from such the angle. Thank you for sharing and good luck!

Thanks a lot, I really appreciate that!
Yeah, once I started looking at AI as more of a support tool instead of a full replacement, everything changed. It’s crazy how much time it saves while still letting you keep the human touch where it matters. Glad you found it interesting!

Good luck, i follow your jouerney.

Thank you bro! I’m glad you’re here for the ride, there’s still a lot to learn and experiment with, but it’s been a fun grind so far. Hopefully the next updates bring even better stuff to share!

 

[shijin]

Day 46: A Minor Glitch and a Needed Reset

If yesterday was calm and productive, today was the opposite — one of those “everything works… until it doesn’t” kind of days.
Nothing catastrophic, just a handful of small issues piling up at once.

Revenue
Finished the day at $462.
Still solid, but slightly down from the recent streak. Most of it came from VIP renewals and a couple of upsells.
No new big tippers today, which is fine — it’s just part of the natural rhythm.

The Glitch
Meta decided to randomly disapprove one of my top-performing ad sets for “misleading content.”
It was literally the same creative I’d been running for a week without problems.
I appealed it, but we all know how that usually goes… so I rebuilt the ad from scratch with slight text changes. Hopefully it’s back online tomorrow.

That pause definitely slowed down new lead flow today — it’s crazy how much a single paused ad can impact momentum once your funnel’s running smoothly.

‍ Models
Model #1 – Had to re-edit a clip because the background music triggered a copyright warning (second time this week). I swear TikTok is allergic to trending sounds.
Model #2 – Her new “morning vlog” reel performed well — 38k views in 12 hours, not bad at all.
Model #3 – Got tired today, said she’s feeling a little burned out from daily filming. I told her to take a breather — no point pushing too hard.
Model #4 – Quiet day, but her follower count keeps rising slowly and consistently.

AI & Chat Flow
Had to fix a minor issue where the AI started looping old responses (again).
Basically, it got “stuck” on a certain phrase and reused it 3 times in the same convo. Not a big deal, but I caught it early thanks to my chatter.
Tweaked the randomness slider a bit and it’s already behaving better.

Still amazed by how much time the AI saves overall. Even when it messes up, the fix is faster than starting from scratch manually.

Traffic
TikTok’s steady — no viral hits today, but engagement stayed consistent.
Reddit was quiet too, which might just be timing.
One interesting note: an older post from last week started getting fresh comments again, bringing in a slow trickle of new followers. The long tail effect is real.

Reflections
Today was a reminder that even when the system feels dialed in, you can’t just coast. Platforms shift, algorithms glitch, and random things break for no reason.
But that’s also what makes the wins satisfying — because they’re earned through all the messy days in between.

I used to stress out over every dip, but now it just feels like… part of the game.

Tomorrow
- Relaunch the paused Meta ad with new copy
- Help Model #3 take a short break + plan lighter content
- Tighten up AI chat filters (so it stops recycling lines )
- Start planning the Month 2 recap post — want to include actual numbers this time

Some days test your patience more than your creativity — but those are usually the ones where you end up improving the system the most.
Onward.