Should You Stop Teaching to Get Clients? [Friday Wrap-Up]

Hey everybody, and welcome to the Friday wrap up on Streamline Solopreneur,

I'm your host, Joe Casabona, and here's what's on my mind for September 4th, 2026.

Number one is teaching versus shifting beliefs.

The thing that's been dominating my thoughts this week is a video from Steve Thompson

called Don't Teach on YouTube. Do this instead to make your first a hundred thousand

dollars.

If I'm being honest, it's been a rough summer for the business. I've heard a chorus

of, yeah, me too. And it sounds like it's the whole industry and

it's nice that it's

not just me or, or so it seems. But that doesn't solve my problem, and it kind of

just makes me feel bad for everybody else as well.

If you've been on my mailing list during the summer, you may have noticed that I've

been experimenting more with content and with offers and with the format of the newsletter.

But at the core of that has always been my belief

that if I teach people how to do something,

the ones who don't want to do it themselves will hire me.

That belief has been greatly challenged this year, and Thompson basically says, why

in this video that I'm sharing

that if you give people the solution,

they'll think they can do it themselves,

But when they get stuck,

they're not gonna get stuck during your business hours and you're not going to be

there to help them.

It's true that I've largely and often attracted a DIY crowd, but I've always hoped

that

in attracting the DIY crowd, I would also get people who didn't want to DIY.

So I can't stop thinking about this

for a couple of reasons, right? I, first of all, I

love teaching.

I've taught in the classroom,

I've spoken many times, I've given workshops. I love teaching people,

but teaching has not yielded new business this year.

And that's been tough. And in fact, if I think back

over

the clients I got over the last year,

they aren't people who came to me

and found me via my teaching.

They are people who googled the need for a podcast coach or a systems coach,

or were referred to me by somebody.

So not only has teaching not gotten me clients, but all of the clients I have gotten

were not

people who found, who saw me teaching and were like, yes, do it for me.

So this really, I know anecdotal, but it really reinforces what Thompson is saying

in this video

that teaching is not the thing that actually gets you the clients.

What, and what he says is, you need to shift someone's beliefs,

and once you shift their beliefs, you become their guide.

And I don't know if he, I forget if he uses the actual term guide, but that's the

term I'm using.

So that's a b

is it seems

completely counterintuitive to me to not make content that teaches something like,

I, like when I watched slash listened to this video on my walk,

I

was confused. I was like, I, what would I,

what

content would I make

that doesn't teach something? And I think I took this a little too literally.

'cause he's not saying don't teach it all. Obviously, you need to teach the belief

system.

What he's saying is don't teach the tactics,

right? Don't do a straight up tutorial on how to do something.

And I'll talk more about this in the something to help your systems, but I've decided

to make September an experimental month.

I want to publish a few videos a weekly. I wanna keep up my weekly cadence

in the format that Thompson recommends,

where I stop teaching as much

and I try to get more paid clients.

Now, this also means big changes will be coming to my mailing list. If you're on

my mailing list, you'll know that

I had a short sale sequence this week.

No one bought from that sale sequence.

I've always been scared of selling too much.

And halfway through this sequence,

I decided I'm going to get rid of that fear.

And I'm not really going, I wouldn't call it scorched earth, because there still

needs to be value in the newsletter.

but I have not been selling,

and I need to get over that fear. My friend Kat,

shout out, Kat, if you're listening,

shared a response someone else had when they were getting complaints about selling

to their list.

And that response is, sorry that I'm trying to do my job.

I like helping people. I really do. And I've always believed that if I just help

people enough, they'll hire me. And that's not, that has not been true this year.

It may not have been true for most of my career,

and I've just been getting false positives because I've been helping people and getting

clients.

But that doesn't mean that one caused the other.

And I'm realizing thanks to this video and some other things I've been thinking about

this week, that the best way for me to help people is by getting paid to do it.

And so I need to do everything I can to get paid to do it,

at least for now. Maybe one day I'll become absurdly wealthy and I'll just teach

for free.

I would love that.

But right now, I, I need to do what's best for my business and my family.

Let me know what you think about this. Do you have something on your mind? Let me

know over@streamlinedfeedback.com. You can write in or leave a voice note. I might

even play it on the show.

Alright, let's move on to recommended reading. This week. That recommended reading

is

three Mount Shasta Climbers rescued after relying on AI to plan climb. This is from

K ctv.com, and I would file this under strong contestants for the Darwin Awards.

I usually focus on the more cerebral dangers of ai. The idea that letting AI do the

work for you will lead to brain rot. You'll lose the ability to think critically

or articulate thoughts without first running to a large language model to tell you

how you feel. And that is extremely problematic. It's the root of everything I talk

about with conscientious AI usage.

But this story shows that there is also a straight up

physical danger.

These people

asked

some large language model,

how to plan their climb.

And instead of,

taking the recommendations and validating them

against some expert,

or

maybe just talking

to an expert,

they decided to just take

what it said verbatim.

the best part of it is the quotes at the bottom. So here's exactly from the article.

I'm gonna quote.

the group said it packed supplies for an eight hour climb, not its roughly 16 hour

ascent.

The climber said, AI recommended simple carbohydrates instead of fats because fats

take too long to digest.

And then they said, we relied too much on AI rather than our own critical thinking.

The group told rangers,

the article, just so you know, the article

says, official said climbers should use reliable sources, carry paper maps and other

backup navigation tools, set a turnaround time and follow it.

They also urged climbers to bring enough food, water, warm clothes, and emergency

supplies for a long day. And an unexpected night

on the mountain. I can only imagine what the prompt was like,

Hey, we're,

you know, we're gonna do a hike. We expect it to be eight hours. What should we bring?

and so like, I, I know nothing about hiking.

And so what I would do in this situation is ask my brother,

or my brother-in-law, both of whom

hike regularly,

my brother would probably be a little more mean about it, or maybe a little more

direct and be like, I don't know if you could do an eight hour hike.

so we rely too much on AI rather than our own. Critical thinking is, I think, an

incredible understatement.

It's an even stronger argument that AI is a word calculator that doesn't quote unquote

know

anything.

And,

possibly because of this story this week, I also came across the AI mushroom meme.

I, this is, I think this was going around in July, but it's basically a person holding

up a mushroom,

to an AI avatar and saying like, Hey, can I eat this? And it's like, yes, absolutely

you can.

Uh, and then the next panel is, uh,

a gravestone and it says, you're absolutely right, that mushroom was poisonous. Would

you like to learn more about poisonous mushrooms?

So,

I came across that meme this week and an article arguing that the meme is propaganda

stating that AI must never be wrong.

Uh, I think the article felt a little ramly at times, like it mentioned, like childhood

vaccines at some point.

and this author who I am going to bestow the dubious honor of being AI blue pilled

on,

uh, misses the point of the meme. The meme is not that AI must always be wrong.

The point of the meme is that we shouldn't blindly trust ai.

So what should these people have done? Well, I kind of mentioned it already, use

actual trustworthy sources and people with experience would be number one.

But if you are going to turn to ai,

use it to surface trustworthy sources,

You say, I am thinking about doing this.

Can you

surface

advice and cite primary sources

for the, for what I should do? Right?

And then you can go and vet and, you know, maybe something from,

you know, the, the National Park Service is going to be better than some rando on

Reddit.

So AI is a fast moving field. We, we need AI literacy.

It's going to be hard to figure out,

you know, what, what AI literacy today is versus a year from now. But using AI is

axiomatic of using or reading pretty much anything on the internet. It's the same

lesson that we learned from blogs and social media that just because it's printed

on the screen does not make it true.

we need better, we need to do better,

uh, because these people very well could have died

because they took advice from AI

instead of some trustworthy source.

Man, what a, what a weird way to go

to go from this to something fun.

but that's, that's the format I'm sticking with.

this something fun is DJ Earworm mashups.

I've been listening to their mashups for over 15 years now, and this week I rediscovered

some of my favorites, including the one,

uh, I'm linking in the show notes party on the floor from summer 2011.

they have such a unique approach to creating mashups and accompanying and, and the

accompanying videos,

which are really good, and I love them,

as someone in his forties, the newer mashups also let me know what songs are popular

these days. So recommend DJ Earworm. Have a, listen. There's a lot of really good

stuff on their channel.

Okay, let's round out with something to help your systems.

I'm calling this the video outliner. So,

related to,

the on my Mind segment, I decided if I was gonna try Steve Thompson's approach to

YouTube, I needed something that would keep me honest using Claude code. I built

a video outliner in H-T-M-L-C-S-S and JavaScript.

So if you wanna see a screenshot of this, I do include it in my newsletter.

I, I can also show it on screen on the YouTube video or the video, I guess, depending

on where you're listening slash watching.

so I took several of Thompson's videos, videos that I watched, not just videos I

blindly agreed with,

uh, or blindly agreed with the analysis of.

Uh, and I fed them into Claude asking it to review the common themes and format.

It identified an arc with clearly labeled sections, and then Thompson's common answers

to those sections. So like, who's the enemy and what is the advice of the enemy once

the symptom, that sort of thing. So I took that, I made a table and I filled in my

equivalent, which also live on a sticky post-it note on my wall.

but I, I filled in this table and then I gave all of that analysis my table and the,

format I like for creating videos, which is story, problem, solution, call to action.

and I had Claude build a Mad Libs style page for me, which again, you can see as

a screenshot in the show notes,

or up on the screen in, in the video if you decide to watch the video.

Uh, but again, I'll include a screenshot in the, in the show notes as well.

The very crucial part here

is that this gives me the beats, but never the words. I write the entire outline

slash script using this format. And the reason that I decided to do it this way

is because as I mentioned earlier, this is so different from my normal approach to

creating videos that I wanted to make sure I was actually sticking to this format.

And I think like

using a large language model to build something

around these guardrails is, is a very good use of it, right? It's doing the task

and not the craft. I am still doing the craft, and I'm just following the requirements

I've set forth for myself.

An important part of

this experiment

is consistency,

and this little application will help me stay consistent.

So that's it for this episode. These episodes are about taking more time off, especially

on the weekends. And the best way you can do that is by having a good system in place

to stop you from doing everything yourself. Now, if you are not sure where to start,

I am offering free 25 minute coaching calls to listeners. This will be one place

where I actually do teach a little bit.

You can book yours over at streamlined FM slash clarity, streamlined FM slash clarity.

Thanks so much for listening to the Friday wrap up on Streamline Solopreneur. And

until next time, I hope you find some space in your weekend.