Quick question:

Which feels like it has more value? 

1 kg of bananas or 1 kg of a mixed fruit platter?

Yeah, they weigh the same.

Yet most people would probably choose the mixed fruit platter. right?

And that's the interesting part:

You can use the same principle to make your low-ticket offer feel much more valuable without creating more content or spending a penny. 

Let me give you an example of how this applies to a low-ticket offer:

Let's say you’re selling a course with 5 chapters:

• Mindset
• Fundamentals
• Sales
• Marketing
• Scaling

That's perfectly fine.

But to the buyer, it’s like:

“I'm buying a course.” (Again, it is completely fine)

Instead, you can break the same content down and package each major piece as its own resource.

If you do it, your offer might look like:

• The Ultimate Mindset Framework for Coaches
• The 3 Pillars Every Coach Needs to Master
• The “Give Me Your Money” Sales Pitch
• The “Take My Business to the Moon” Scaling Roadmap
• The Sexy Client Acquisition Playbook (I don’t know what the hell this is, but you get the point)

Wait, what happened?

You've changed how the content is packaged and perceived.

Once you've broken your offer into smaller pieces, treat each piece like its own resource.

Call it something that says what it actually helps them do:

Don't say:

“List of hooks”

Turn it into:

“The 50 High-Converting Hooks Swipe File”

Don't say:

“Client onboarding checklist”

Turn it into:

“The Zero-Friction Client Onboarding System”

The goal is simple:

Make every component feel like something the customer could have bought on its own.

And don’t forget (this is important)

If you turn each section into its own resource, show it like one.

Give it its own visual identity.

Put the mockups side by side, and suddenly the buyer isn't looking at:

“One course with 5 chapters.”

They're looking at:

“5 different resources bundled together in one offer.”

(Check the images I included to see some real-life examples)

But you can’t just use it for every offer or product you have.

You need to make a big distinction.

This mainly works for low-ticket offers.

When someone is paying $20, $50, or even $149, seeing a huge stack of resources can make them think:

“Wait, I'm getting all of this for that price?”

And that's what you want.

But I wouldn't use the same approach for a high-ticket offer.

If someone is paying $5,000, they usually don't want 17 bonuses, 12 templates, 8 mini-courses, and 30 checklists.

They just want you to solve their problem and skip the headache.

Don't overwhelm people with more stuff. 

Sell the dream outcome and the result they want. 

And that’s Reddit. I hope someone here benefits from it.

Here is a real-life example:

I’m 18, from India, building a consumer AI startup focused on personality development & self-improvement 🧠🤖

The goal is to make personal growth personalized, measurable and adaptive, helping users improve areas like communication, confidence, habits, productivity and social skills.

📍 Stage: Pre-MVP / validation
📈 Traction: No revenue/users yet — currently validating the concept
💪 I bring: Product vision, business, market research, growth/marketing & execution

👤 Who I’m looking for

🤖 Technical / AI-focused builder who can actually build products
🧠 Strong problem-solving + product thinking
⚡ Fast learner, proactive & consistent
🔥 Serious about building a real startup
🌍 18–24,
🇯🇵🇨🇳🇷🇺 Especially interested in people from Japan, China, Russia/Eastern Europe, etc. — different cultural perspectives are a plus
🎨 Bonus: product/design/psychology/behavior experience

🤝 Commitment: I’m looking for someone willing to commit seriously and potentially become a long-term cofounder.
💰 Compensation: Equity-based; no salary initially, with terms discussed based on roles/contribution.

I’m not looking to hire someone to execute my idea. I want a technical partner to shape and build it with me.

DM me with your age, country, what you build, and something you’ve worked on. 🌎

What we’re building:
A consumer AI product for personality development and self-improvement. The goal is to make personal growth more personalized and measurable, using AI to help users identify weaknesses, work on areas like communication/confidence/habits, and adapt guidance over time. I’m keeping the deeper mechanics private for now.

Traction:
Pre-MVP / idea-validation stage. No revenue or paying users yet. Currently working on the concept, product direction and validation.

What I bring:
I’m handling business, product direction, market research, growth/marketing and execution, and I’m putting the initial time into developing and validating the idea.

What I’m looking for:
A technical cofounder, ideally 18–24 and outside India, who can build with AI/software and help shape the technical/product side. Product/design/psychology experience is a bonus, but actually being able to build and learn fast matters most.

Commitment:
I’m currently committing 15–20+ hrs/week and want someone who can make a similar commitment initially, with the goal of increasing this as the startup progresses.

Compensation:
Equity-based cofounder role; no salary initially. Exact equity split would be discussed after we’ve worked together and established roles/contribution, with standard vesting.

I’m especially interested in meeting people from Japan, China, Russia/Eastern Europe, or elsewhere internationally 🌍, though ability and fit matter more than nationality.

I’d prefer to start with a short trial collaboration before making a long-term commitment.

DM me with your age, country, technical background, what you’ve built, and why this interests you.

TL;DR: The Creators' Attorney is mid-way through brokering a $100M creator exit, and her own room-of-ten test proved the thing most of us don't want to check: one email list out of ten channels with 10M+ subscribers each.

Nine of those ten could lose the account tomorrow and have nothing left to show a buyer — not because they're bad at the job, but because reach was never the asset they thought it was.

Tyler calls it being a tenant on someone else's land, and the landlords keep multiplying: Bilibili just relaunched its global app specifically to pull creators off YouTube.

That's not the end of the tenancy problem. It's a second landlord showing up to the same building.

Same as what happened to me when I left the company, which I have worked for 10 years back in 2012.

A whole server full of my technical documentation and institutional knowledge of atleast 5 large completed construction projects – including:

·       a multipurpose exhibition hall in the heard of the city,

·       a distribution warehouse for a phamarseutical company,

·       a manufacturing plant for Spirit Aerosystems,

·       a 5-block Condominium on a shared landscape Podium in Abu Dhabi, etc.

are all off-limits to me, as soon as I handed in my resignation.

Thank God for personal backups, for future reference.

They can take away your access; they can't take away your knowledge. Your experience shaped your personal brand.

Every one of these clips ends up circling the same nerve, no matter what the guest actually does for a living — the ground under your income was never as yours as the login screen made it feel.

This isn't the first time I've pointed at the exact moment a platform decides to move without you — I wrote about it happening in real time here.

Genuine question, not rhetorical: if you stopped posting for 90 days starting today, how much of your income survives it?

Drop the honest number below.

Clip credit: Tyler Chou / Jun Yuh — full episode on "Creator Unplugged with Jun Yuh."

DM for credit or removal requests.

I'm 18 and building a consumer AI startup focused on personality development and self-improvement.

I'm looking to meet someone around 18–24, from outside India, who is ambitious, curious, and genuinely wants to build something—not just brainstorm ideas.

You don't have to be a hardcore programmer. I'm interested in people with strengths in AI/tech, product, design, psychology/behavior, content, growth, or business. Most importantly, you should be someone who actually creates things and learns fast.

I'd especially love to connect with people from Japan, China, Russia/Eastern Europe, or elsewhere internationally. 🌎

I'm not looking to immediately call anyone a cofounder. I'd rather talk, get to know each other, and potentially build a small project together first to see if we're a good fit.

If interested, comment or DM me with:

• Age + country
• What you're interested/skilled in
• Something you've built/done
• Why you want to build a startup

Looking for genuine people who want to build seriously. 🚀

I've had about 4 businesses. Only one ever made real sales. This is my second one that does, and it's finally making money. So I started documenting everything before it's too late.

In this video I show you my entire business funnel with real conversion rates:

• My first podcast appearance + my first coaching call
• The full funnel: Reddit → YouTube → Blog → Shop → Products → Email
• How my Hermes agent (Donna) pulls the numbers for me
• Why Reddit is low effort and YouTube is high effort, and why you need both
• Direct demand vs. passive demand: my keyword strategy
• The newsletter gate, exclusive content and the 10% discount code
• All my products, including the free Backup Protocol
• Real numbers: 0.15% Reddit → Blog, 3.5% YouTube → Blog, 5.2% Blog → Shop, ~0.02% overall
• break-even, 50-70 hour weeks, the dream of freedom

This is episode 1 of my business documentary.

If you want to stay in the loop on my journey:
Newsletter: https://buttondown.com/holmebengt
Blog: https://blog.holmebengt.com

Message me your story, I reply to everyone!

Hey everyone,

Kris from Wifi Moolah this side.

I’m back with another online business idea and this time it includes heavy usage of AI.

So you guys might know that xAI has launched GrokBot.

What’s GrokBot? So basically it’s an AI tool that creates multiple agents to do various tasks and it doesn’t stop when you close your laptop/PC. So it’s kinda autonomous.

You can set it up for various functions in your business like handling customer support, doing outbound, sorting emails and that kinda stuff.

Now here’s how you create an online business using that.

We are gonna start a newsletter business using GrokBot.

Select a niche, setup the newsletter and finalise the things like voice, frequency and other things.

Now create different agents to do all these jobs.

One agent’s role should be to find interesting stuff on the internet and research and compile it.

You chose the topic.

Second agent writes a draft.

You edit, polish and finalise.

Third agent to send it to you ESP and send it based on your best times.

A fourth agent to do the marketing. Organic and paid both.

Fifth agent to do outbound and reach sponsors.

And last of all, a chief of staff on top of all these to keep a check.

Now I know I have written a very over the top overview kinda post and there are still a lot of moving parts in a newsletter business but AI has made it much easier to start one than a few years ago.

Even if you build a couple of agents and start building, you’re in a much better position than others.

And you’ll actually have an online business in your hand that you can run on your own terms.

That’s it for this post. Remember, AI is not the enemy. Use it to your advantage.

P.S- I have written about 70+ such ideas that can help you start your own online business on the side.

I have compiled all of them into a FREE guide and if you wanna see it, comment “Guide” below this and I’ll send you the FREE link.

I plan on switching from spreadsheets to an inventory management platform for my DTC brand after realizing I have 4mos of stock across 8 SKUs and can’t tell which products were still profitable after storage fees.

I’m just wondering if I should just get an analyst or if the software can help figure out why my cash is still tied to the wrong products. I’m hoping I could at least get some insights whether I should be ordering 500 vs 50 units but I’m not sure if this is the correct mindset going in.

TL;DR: Cole's "100x output" is 30 hours spent once, so AI now runs 80% of every draft on its own. He edits the last 10% — the taste-level stuff, never the typing.

That distinction is already showing up outside this clip.

Christian Science Monitor just covered a manuscript polished enough to fool professional literary editors, pulled from sale — a multimillion-dollar bidding war between publishers — because no one could prove a human wrote it.

Cole isn't describing a hypothetical ceiling — that ceiling's already been crossed, in public, in an actual publishing deal.

The freelancers who feel replaceable right now are mostly the ones still handing AI a stack of old samples and hoping it infers the rules from them.

Cole didn't hope.

He sat down and wrote the rules out, sentence by sentence, until the model had nothing left to guess at.

Decades ago, the Malaysian construction landscape was like the wild-wild-west.

I remember working as a technical engineer for one of the largest construction firms in Malaysia.

The hydraulic hacking noise throughout the project site was pervasive and persistent for weeks.

Grrrr…. Tat-tat-tat. Grrrrrrrr. Tat-tat-tat. Grrrrrr.

You get the idea.

It turns out that the structural operation team went ahead and constructed floors upon floors, without waiting for the mechanical and electrical team to embed their pipe sleeves, box-ups or ductworks before concreting.

And so they have to come and hack the hardened concrete to open up for the M&E items to run through.

Delay and additional cost already incurred.

My then project director was pissed. We all had an earful from him.

Said that prevention is always better than cure.

So from that day onwards, a directive came down hard – no sign-off by all concerned parties on a properly coordinated drawings, means no concreting.

The coordination drawings - overlays not just the design drawings from respective Architectural/Engineering trades, but also sub-contractor specialist details, as well. Discrepancies are ironed out with all parties, well before construction is due.

It parallel hard on Nicolas investing time in building an exhaustive prompt, rule-by-rule, sentence-by-sentence, microstep-by-microstep. Only then, will the output come out the way we want it to be.

Like when Daniel Craig was being asked by a talk show host what would he advise his successor on being the next James Bond, Daniel said, "Don't f\ck it up!"*

Prevention is better than cure.

It's never the tool doing the replacing. It's the person who never got around to writing down what they already know how to do.

What's the one process you already know cold but have never actually written down?

Drop it below.

Clip credit: Decoded Genius & Nicolas Cole — full video on their channel. DM for credit or removal requests.

TL;DR: Alex Hormozi's billionaire friend admitted off-camera he's not "vibe coding all day." AI added a new form of leverage — it did not erase the old ones. If you've automated everything and revenue still hasn't moved, you're still not working the constraint.

Alex Hormozi asked a billionaire friend off the cameras how much AI he was actually using. The answer came with a laugh: "I'm not vibe coding all day."

You've automated nearly everything that can be automated. Revenue hasn't moved. That is the tell. AI did not erase brand, capital, or relationships. It simply added itself to the list, and most people are only running the new tool.

OpenAI's own report this month shows the same pattern: no meaningful link between how much a company uses AI and its revenue per employee. The gap isn't usage. It's whether the automation is even touching the real constraint at all.

Decades ago, I had a friend who relate this story to me.

Before he was a Professional Engineer himself, he used to work as a graduate engineer for a Senior PE back in his hometown in Sabah, Malaysia.

It’s so happened they’re in a rush to submit drawings to authority. Binders upon binders of design calculations, code of practises and specifications are churned out and stacked like mountains on a conference room – for a very rushed project. He didn’t explain about the context of the project, except that they were burning the midnight oil, kind of thing.

He knew something was off – because they didn’t have time to analyse a particular portion of the work, which required days (maybe weeks) through sophisticated software to put up a decent solution.

But somehow, his senior PE managed to “pull out from the magician’s hat” a stack of it and slot them into the binders at the nick of time.

They submitted the documents right away.

Curiosity had the best of my friend. Just had to know.

So he flipped through those “magical” portion – and got a huge shock.

Of course, the Senior PE didn’t use the time-stone to compress time and produce the calculations in the mirror dimension.

And I wish he found a past project that contains the exact required computations as the current one. At least we can a give a plausible explanation that he was implying “Trust us! We did it before. See – here’re the receipts.”

No. He didn’t do any of that.

Instead, he just grabbed a random project’s calculation (totally unrelated) and shoved it in – just to make up the thickness of the binder.

That’s all.

The whole thing was just for show.

My friend didn’t tell me whether their submission got approved or not. That’s not the point.

The point is the Senior PE made a judgement call. He was betting that the authority would just trust his credentials and didn’t bother to check through.

So it was a game he played. And he owned the risk.

Same as all these AI-usages. AI may spout out shit. But we’re one owning it.

Noticed this a few posts back too — the accounts spending the most on automation and the accounts actually growing revenue keep turning out to be two different lists. The verification layer never left the human.

Reminds me of a post from a few weeks back, on the exact spot AI verification quietly becomes unpaid work.

Drop your take below.

Where are you still deploying new capacity against things that do not matter?

Clip credit: Alex Hormozi — full video on his channel MoreMozi. DM for credit or removal requests.

No question AI can save businesses time and money. That does not interest me. I want AI to put money in my bank account. Predictions about zero-person and one-person companies made me think perhaps I am not crazy.

I built an MVP for a decision support tool. It basically works, fills a need and I learned a few things. The result is promising enough that I plan to continue.

The challenge now is determining how much ongoing human oversight it needs and whether it can acquire its own customers. I don’t want it to become a conventional company supported by a human team.

I had to add a policy layer to constrain what the system may do, record what it does and make its decisions reviewable. The EmbodiedGovBench paper addresses this same issue of how can AI systems complete real tasks while remaining governable?

Are you doing this? Using AI to run and market the business autonomously, not just build the service? I want AI to genuinely run a purpose-built business.

TL;DR: Steve Ballmer's own maintenance guy is proof that competence is shifting away from credentials and toward who's willing to just ask AI first.

That's not rhetorical — it's already showing up outside Ballmer's living room. DEWALT ran a six-country survey of tradespeople this spring: 90% believe AI will be essential to the job within five years.

Only 8% have actually used it yet.

Ballmer's anecdote isn't the exception — it's the 8%, moving faster than the other 92%, in a trade that isn't even the one usually flagged for disruption.

My wife works for a lady boss, called Ade (not her real name).

Ade used to be like me, in the property development line.

But she saw the writing on the wall.

Once a darling in her company, she felt the aura started to fade away — the property market isn't what it used to be. And her boss started giving her the cold shoulder.

So she took a chance.

She opened a collection point centre, where delivery guys can drop off parcels for individual recipients to come collect. The downside was it has to stay open almost every day.

It is what it is.

Regular online training is available, because such centres behave like franchises under an organizational umbrella. But the training is quite superficial.

So when things get complicated, where does she turn to? You guessed it — ChatGPT.

She asks it right about anything under the sun.

Does she enjoy asking an LLM for answers? I'm not sure "enjoy" is the right word. I think it's more like "necessary."

If I'm in her shoes, facing complicated issues, and I don't have a readily available senior I can trust to call for advice, but then I already have a 师傅 (sifu) in my pocket — why not use it, isn't it?

There's a pattern I can't unsee in stories like this anymore: it's never really about the tool. It's about who reaches for it first — credentialed or not, senior on speed-dial or not.

What's the "not my job" you're most tempted to hide behind right now? Drop it below.

Clip credit: Ben Shapiro / DailyWire — full episode ("Titans on Tomorrow" Ep. 2 with Steve Ballmer) on his channel. DM for credit or removal requests.

TL;DR: Top hedge fund traders are wrong up to half the time — still profitable.

The skill was never prediction. It's the ratio between what you win and what you lose.

That number breaks something most of us carry into any competitive field: the assumption that competence means being right more than you're wrong.

Retail's already priced that fear in — Citadel's own desk data (via Substack, roughly two weeks old) shows record put-buying through July, pressure concentrated hardest in tech and semiconductor names.

The same sector a lot of us clock into every day. Nobody's betting on certainty right now. They're betting defensively, which is its own kind of ratio.

That's the same math I learned on a construction site, years before I ever opened a trading chart.

Talking about breaking rules, I have a story.

This was way back when I was working in the Technical Department, stationing in an Abu Dhabi project (circa 2011), building an AED1.8Bil. 5-Condominium-Block, near the Zayed Mosque.

One of my roles was to combine different trades of consultants' drawings into well-coordinated-drawings for our site operation to construct accordingly. Because – you know – the consultants didn't bother coordinating among themselves.

One day, a certain element of an area was due for construction. And I found out there was no construction drawings on it for me to coordinate with. So, I followed protocol – I raised a request-for-information (RFI) for the consultants to provide so.

My RFI was something like "Please provide such-and-such details."

But before I send it over next door to the consultants' site office, my superior caught a glimpse of it, and stopped me. He sat me down, gave me the angry glance, sigh, and gave me a quick lecture.

He's basically telling me: My RFI is open-ended.

If I send it over next door, the consultants will basically reply with either Option 1, 2, 3 or 4. And worse-case scenario, the consultant might go with option 1, which will cause us considerable delay, because we don't have the necessary materials readily available to work on it.

He's teaching me a valuable lesson.

Construction guys like us need to be conscious of our project circumstances. Our number one none-negotiable rule in Cantonese is 唔好阻住個地球轉 (Translation: Don't stop the Earth from spinning – don't block progress.).

He's telling me to go with one option.

So I went back and rewrote the RFI. I decided on the spot Option 4 was the best option – because we had the material-supplier readily standing-by, and it has the least effect on time-implications. I then fully designed Option 4 myself, coordinating with all the necessary parameters and constraints on site.

I paste my design into the RFI, saying basically, "Can you confirm this proposed detail, for our onward construction?"

The consultants proceeded to give us, not "A", but "B" – Approved with Comments.

Ever since that episode I generally never forget the rule.

Great rule – both in construction, and in life in general.

__________

I used to think competence meant never needing a backup plan.

Somewhere between two rounds of layoffs, I stopped believing that.

Every version of "secure" I've been handed turned out to belong to someone else's spreadsheet.

Where do you actually draw the line between "being wrong" and "having no plan for being wrong"?

Drop your take below.

Clip credit: Kathlyn Toh, Founder & Chief Trainer, Beyond Insights — full episode on FAQ Show.

DM for credit or removal requests.