Full Transcript

·YouTLDR

If I Wanted to Build an AI SaaS in 2026, I’d Do This

28:39EnglishBy Dave EbbelaarTranscribed Nov 17, 2026
Analyze another video with Pro30-day money-back guarantee
0:00

So, building a B2B AI SaaS is probably

0:02

one of the hottest things you can do in

0:03

tech right now. But, most developers

0:05

that actually get started with this got

0:07

it completely wrong, completely in

0:09

reverse. They go into their cave, work

0:11

on something for 6 months that they

0:12

think is a great product, and then they

0:14

figure out that actually nobody wants

0:16

it. So, in this video, I'm going to show

0:18

you the path that I would recommend if I

0:20

were to start over from scratch today

0:22

and build a new software company from

0:23

the ground up. It's the path where you

0:25

actually get paid first and then get

0:27

actual proof before you call it a

0:28

product. And now, this is the exact same

0:30

methods that I've been using for the

0:32

past years already to build and launch

0:34

multiple software companies. And there's

0:36

going to be three steps to this. And

0:39

most people actually go straight to step

0:41

number three, which is where all the

0:42

problems come from. So, in this video,

0:44

I'm going to show you how you can safely

0:46

go from step one to step two and then to

0:49

step three in order to bootstrap a

0:51

software company from the ground up on

0:53

your own or with a small team without

0:55

relying on external funding, so you keep

0:57

100% of your shares. And a quick

0:59

disclaimer before we dive into things.

1:01

I'm not a software millionaire. I

1:02

haven't made the nine-figure exit or

1:04

built this gigantic unicorn, but over

1:07

the years I've built my fair share of

1:09

software. Some products making a couple

1:11

hundred dollars per month, some a couple

1:13

of thousands, but it all adds up and the

1:15

pattern that I've used throughout the

1:17

years is always exactly the same. And

1:19

now, what you should really understand

1:21

about building software is that the

1:23

hardest part of this whole business game

1:25

was never

1:27

the code. This was never the hard part.

1:30

Now, more than ever with AI coding

1:32

agents, the code is the easy part. The

1:35

whole game of software has always been

1:38

about finding a customer, and in

1:41

particular your first one, this is the

1:42

most important one, plus getting them to

1:45

pay, plus getting them, most important

1:48

part, to stay. This is the whole game of

1:51

software. Where most people go wrong,

1:54

especially engineers, technical

1:56

founders, they start with the code. So,

1:59

they start here. And then they start

2:01

building something for 6 months and

2:03

figure out they cannot find any single

2:06

customer. They're not willing to pay,

2:07

and even if they find someone to pay for

2:09

it, they don't stay. The beauty of

2:11

software is that if you get this right,

2:13

this loop of customers paying and

2:15

staying, you have what they call product

2:19

market fit. In the beginning, this is

2:22

the most important thing to work to

2:24

because if you can find customers, they

2:26

pay and they stay, you have a beautiful

2:28

business model, and that's why software

2:30

companies are so attractive. Okay, so

2:32

then how do you get to product market

2:34

fit? Well, if you don't start with the

2:37

code, what do you do, right? Well, we

2:39

flip the script and we start with the

2:41

customer. And then you end up on what I

2:43

call the surface-to-software

2:45

ladder. So, this is the model that I've

2:48

been using where you start in a very

2:50

unsexy and unscalable way to building

2:53

software, really. And that is you start

2:55

as a freelancer or as an agency, or

2:57

simply just as a developer who built

2:58

something for a particular company, and

3:01

they pay you to do the build. So,

3:03

there's no fancy sign-ups, no fancy

3:05

multi-tenant support. You simply find

3:07

one customer with a particular problem

3:09

that you can create a build for, and you

3:12

let the customer pay you for that build

3:14

so you can prove your concept first. So,

3:16

that is the overarching idea of the

3:18

surface-to-software ladder. So, now

3:20

let's get into step one to explain how

3:23

to actually get going with this. And

3:25

step number one is what I call the done

3:28

for them phase. And this is simply where

3:32

you do all the work. This is where the

3:34

surface part comes in. And the whole

3:36

goal here is to find that one particular

3:39

company that has this problem, and let's

3:42

zoom in a little bit on how we can get

3:43

started here, right? Because this is

3:45

also the hardest part to get right. So,

3:47

you might have questions, where do I

3:48

start, how do I find that first client,

3:50

what problem do I actually solve? And

3:53

the word that you want to remember and

3:55

that you want to optimize for is what I

3:57

call breakage. And breakage in a

4:00

business can be any type of

4:03

problem leakage where they are either

4:06

leaving money on the table by not

4:08

solving that particular problem or a

4:10

problem is actually costing them a lot

4:12

of money on a weekly and a monthly

4:15

basis. So, this is what you ultimately

4:17

want to look for. But the only way that

4:19

you can kind of like get to that point

4:21

is by getting insiders knowledge into

4:24

what's actually going on inside the

4:26

company. Because often, this is the very

4:29

the unsexy stuff, the stuff that goes on

4:32

in the background in certain

4:33

departments, certain employees, between

4:35

certain systems. And without insiders

4:39

knowledge, it's usually very hard to

4:41

understand what this is all about. And

4:44

most of the the big software companies,

4:47

they fix already fixed the obvious

4:49

stuff, right? So, people need to send

4:50

emails, people need to manage, people

4:52

need to plan tasks. So, you are not

4:55

going to build the next Gmail or the

4:56

next Slack or the next task manager.

4:59

Because good luck competing with those

5:01

giants, right? So, those are the that is

5:02

the obvious stuff, the obvious breakage.

5:05

Where we want to get to is is way more

5:08

niche and way more subtle. And the only

5:10

really the only way to find this out is

5:12

to actually

5:14

go and talk to people. And I know for

5:16

most engineers, this is the scary part,

5:18

right? This is This is why we hide in

5:20

our caves for 6 months and build stuff.

5:22

But trust me, that is a recipe for

5:24

disaster. You actually need to talk to

5:26

people. And instead of like cold

5:29

emailing or pitching a whole bunch of

5:30

people or going on Upwork, what I

5:32

actually recommend is where to start is

5:34

with your warm network. Pretty much

5:37

anyone that already kind of knows you

5:39

and if you send them a message, they're

5:41

not like, "Who is this person?" They at

5:43

least to some degree, there's some

5:45

familiarity there. And why you want to

5:47

start there first is because if you

5:49

compare that, for example, with cold

5:52

outreach, which is the opposite, right?

5:54

Where you reach out to someone that

5:56

pretty much, let me use some colors over

5:58

here. Cold outreach, and then we have

6:01

warm outreach with your network. The

6:03

difference pretty much is like cold,

6:05

these people have no idea who you are.

6:07

Um, for example, on a network on

6:08

LinkedIn. So,

6:10

for you to get a reply and for someone

6:12

to actually decide, "Okay, let me

6:13

actually take a Zoom call with this

6:15

person or meet in person to talk about

6:17

my business." You're going to have to go

6:20

through a lot of no's in order to

6:22

eventually get one yes. Whereas with

6:24

your warm network, it's way easier to

6:26

get an introduction and to just to just

6:29

have a chat. Because the thing is

6:31

nowadays with AI

6:33

and what AI can solve, there is breakage

6:37

in pretty much every company. Because

6:40

there's so much AI can solve right now

6:42

that wasn't possible

6:44

a couple of years ago. So, you'll find

6:46

all of these businesses that have all of

6:49

these processes like admin, support,

6:54

sales, marketing, like pretty much all

6:57

the departments of a business. What else

7:00

we got in here? We got all the ops, of

7:02

course. Operations. All these areas have

7:05

people and processes and systems

7:09

working on something. And usually, this

7:11

is in a company it like this is all very

7:13

messy. Because over the years, it builds

7:16

up. And now with AI, we can kind of like

7:19

step in. And by the way, this this nice

7:21

circle over here represents the

7:23

complexity of business. Like, I don't

7:25

know. That was just what came to my

7:26

mind. So, now with AI, we can just come

7:29

in and we can try and fix all these Let

7:32

let's see. If we have the

7:34

the different operational layers, if you

7:36

will, in a company, right? You have all

7:38

of these breakage points at different

7:41

levels of the company all relating to

7:43

these different types of departments

7:46

that things fall under. So,

7:48

we can step in and one by one we can

7:50

identify these opportunities. And the

7:54

thing is, you should position yourself

7:56

when you jump on these chats with people

7:58

not as

8:00

"Hey, I'm going to pitch my services and

8:03

I'm going to say, look, I know all about

8:05

AI. I know how to solve your business.

8:07

Here's what I can do." No, instead, you

8:10

position yourself

8:12

as curious. That's probably the best tip

8:16

that I can give you. So, instead of

8:18

being salesy, being a pitch, thinking

8:20

about you thinking that you know what

8:22

they need, you position yourself as

8:25

really curious. And you go through a

8:27

process what's in sales known as

8:31

discovery. And this is simply asking

8:34

questions. Really getting to the core

8:37

problems of what they experience within

8:41

a company. And you need to do some

8:42

digging in there. What you want to start

8:43

listening for are the following signals.

8:46

So, you want something that is kind of

8:48

like described as as painful, ideally

8:51

also repetitive,

8:54

costly, and also unfixed. If I had to

8:57

boil it down, this would be the

9:00

definition of breakage. And if you find

9:05

a process, if you find something within

9:07

the business that checks all of these

9:09

boxes, so both painful, repetitive,

9:11

costly, and unfixed, you have a serious

9:14

opportunity to make a lot of money. What

9:17

you then do is you propose a solution.

9:20

And this can come in a variety of forms,

9:23

but the key thing here is that you want

9:25

to keep it really simple, lean, and

9:28

minimal so you can build it quickly, so

9:30

you can really position this in like the

9:33

software world is what we call as a

9:34

proof proof

9:36

So, you want to build quick, and

9:38

especially nowadays with AI, this gets

9:41

easier and easier and easier to get like

9:43

in a day, 2 days, maybe 3 days, get

9:45

actually something that already works

9:47

and can show some tangible value. So,

9:49

you want to propose a POC, but there is

9:52

also one important part. You also want

9:54

to propose an actual dollar amount to

9:57

it. Because

9:58

everyone wants their problems solved if

10:02

you go out and just do it for them,

10:03

right? If you ask, "Hey, can I build

10:04

this for you to solve this problem?"

10:06

Like, "Yeah, sure. Why not? Go ahead,

10:07

try it. We'll see what comes out of it."

10:09

As soon as money is on the table, then

10:12

the situation and the dynamics switch,

10:14

right? Because now it becomes an

10:15

equation. It becomes a question of is

10:18

what we are putting into this thing, is

10:21

that worth more or less than we are

10:24

getting out of it in terms of value? And

10:26

how you can go about this is you

10:27

actually create a proposal for them. So,

10:30

here's what that could look like. You

10:31

could say, "If I build you X, and then X

10:36

is the the nice solution that you're

10:38

going to propose that solves the problem

10:39

in a very lean and minimal way, and then

10:42

given it solves Y, this can be some type

10:46

of guarantee, meaning you could say if

10:49

it solves a certain percentage of

10:52

problems, if it saves you a certain

10:54

percentage of time, or makes you a

10:56

certain amount of extra revenue on a

10:57

monthly basis." Again, it all depends on

11:00

what you are solving and what the unit

11:02

economics really of that problem are and

11:05

how you can make an ROI ROI calculation.

11:07

You can put a guarantee on that. If I

11:09

build you X, given it solves Y based on

11:12

this guarantee, would you be willing to

11:15

pay me, let's call that Z, for the

11:18

initial build plus another, let's call

11:22

this, I don't know, W

11:24

dollars per month to maintain this for

11:29

you? So, now you get to a proposal. And

11:33

the whole thing, the whole the the the

11:34

true test over here is them willing to

11:38

pay for it. This is so important because

11:42

if they're not willing to pay for it,

11:44

probably other customers are also not

11:46

willing to pay for it, and that means

11:48

the value

11:50

is not right, and that means you don't

11:53

have a product market fit yet. And then

11:56

quick thing, if you're getting value out

11:58

of this video so far, would you do me a

11:59

favor and hit that like button below

12:01

this video, and also consider

12:03

subscribing to the channel if you want

12:04

to see more content like this. That

12:07

would really help me out. Thanks. So

12:08

step number one, doing all for them is

12:10

all about the unsexy, unscalable work.

12:13

You're essentially delivering a white

12:15

cloth surface fully custom to that

12:18

particular client. And you should not

12:20

skip this, and this is also not below

12:22

you. If you look at some of the most

12:24

successful tech companies like Stripe

12:26

and Airbnb, this is exactly how they

12:29

started. Okay, now the beauty of this

12:31

whole phase one is that you actually get

12:34

paid to do this. So you don't have to

12:37

like go into this founder cheap founder

12:39

mode and like work for months on end

12:41

before you get anything. You actually

12:43

get paid to prove your idea, which is

12:45

the best thing about this approach. Now

12:48

before we get to step two, which will be

12:51

up next, there's one more test that you

12:54

need to do in order to figure out if

12:56

you're actually building the right

12:58

thing. And that is the what I call turn

13:01

it off test. Very simple.

13:03

Turn off. So if you have a great product

13:08

that a business relies on, that gives

13:10

them great value,

13:12

if you turn it off, ideally within the

13:15

next hour, you should get a phone call.

13:17

Hey, this is not working. Like what's

13:19

happening? What's going on? The sooner

13:21

that phone will ring and they will start

13:23

to call you wondering what's going on,

13:25

the better your product. So, think about

13:28

for example, a customer service or

13:31

customer support automation system. Go

13:34

this. So, this is actually something we

13:36

worked on a lot. So, customer support

13:38

automation. So, tickets would come in

13:41

into an inbox. So, there was an email

13:43

inbox. This is my best drawing of an

13:47

email inbox. And then, what would happen

13:49

is the AI would start working on this

13:52

and start processing and then would

13:55

reply to the customers, right? And it

13:56

would have an escalation gate to

13:59

escalate to a human support agent if

14:01

things go wrong.

14:02

We've had plenty of times during our

14:05

first like the first initial phase where

14:07

we were onboarding things, sometimes

14:08

something would break or like a model

14:10

wasn't available anymore. Like all of

14:12

the stuff that happens in the beginning,

14:13

right? That support team is there every

14:16

day like going through all these

14:18

tickets. And this was a huge time saver

14:21

for them. So, they only really got about

14:24

25% of the tickets that were escalated

14:27

that the AI couldn't handle. As soon as

14:30

this system was not working anymore, the

14:33

workloads of those what they had to do

14:36

went from like now they don't have to do

14:39

25, they have to do 100% of the tickets

14:42

again. So, if you do the math, that's a

14:45

4x in total work that needs to be done

14:48

with the same team. So, that is how you

14:51

should think about the

14:53

the unique economics and the ROI

14:55

calculations of software. Now, this is a

14:58

very like simple example. It's very

15:00

straightforward, but you ideally always

15:02

want to look for something like this

15:04

where you can get to these types of

15:06

calculations where if they if you turn

15:08

it off within hours, they will call you.

15:11

So, if you get to that point with your

15:13

first build, you're actually in a really

15:15

good spot to make a lot of money and it

15:17

can be very successful. I would even

15:19

make it conditional before going into

15:22

step and phase number two, where

15:24

otherwise I would just continue doing

15:26

step one over and over again until you

15:29

find something that checks all of these

15:30

boxes. Okay, so once you're there, we

15:32

can go to phase number two of the

15:35

surface to software ladder. Let's call

15:37

it step number two, and this is where

15:39

you move from fully done for them, you

15:42

move to done with them. So, this is

15:46

where you're going to do some of the

15:47

work, but you're also going to include

15:50

the client in there. And the whole goal

15:52

of this phase is to get three to five

15:57

near identical customers as your

16:01

original customer. So, the whole trick

16:04

that you're going to do here is you now

16:05

have that product skeleton, right? And

16:08

again, it's still it's POC phase. It's a

16:10

little bit sloppy. It doesn't have all

16:11

the fancy onboarding and the

16:12

multi-tenant support, but you're going

16:14

to another company that's very similar

16:17

and that also has that same breakage,

16:19

that same problem that you can solve.

16:21

And you rebuild it and integrate it

16:24

within their company again. So, you

16:26

don't even necessarily have to onboard

16:28

them to the build you already created.

16:29

You can do another like custom built.

16:32

Because in the beginning, you have to

16:34

figure out look what what is the 80%

16:37

that's the same for everyone, and what's

16:39

the 20% that is custom tailored. So,

16:41

what I recommend here is also when

16:44

you're trying to find that next

16:46

customer, up often the best starting

16:49

point is to literally ask your client

16:51

number one for other companies in their

16:53

industry. Because like people network,

16:56

people know the the industry and the

16:57

branches that they are in, right? So,

16:59

that is the best way to get your foot in

17:02

the door. And I recommend to go as niche

17:06

as possible here. So, with software you

17:10

don't want to be generic, because anyone

17:13

can build a generic task manager. Anyone

17:16

can build a generic marketing app,

17:18

right? But, for example, when we were

17:20

building this customer support

17:22

automation system, we focused on a

17:24

particular niche really. So, we did

17:26

e-commerce stores who were using Gorgias

17:30

as their ticketing system. So, this

17:33

combination was already very specific.

17:35

So, the whole goal here is to find more

17:37

customers, and then you want to slowly

17:41

work towards this, like done with them.

17:44

That's kind of like the key phrase over

17:45

here. So, you do some of the builds, but

17:49

you're also going to build some, let's

17:51

say, simple dashboards that they can

17:54

see, or onboarding steps, or certain

17:57

forms and even settings that they can

18:01

fill in to now start using the software.

18:04

So, you shift more and more towards the

18:07

customer. Now, another thing that you

18:08

want to switch here is the pricing

18:10

model. So, how to think about pricing?

18:13

Whereas in phase one, it was all about

18:15

the initial build cost, and then the

18:17

maintenance fee afterwards, on the

18:20

second and clients after that, you

18:22

ideally want to turn to some types of

18:24

value-based

18:26

pricing, and you don't have to pay as

18:28

much up front anymore on like the custom

18:30

build. So, again, for the example of the

18:34

ticketing system, we, for example, the

18:36

original build, we did like a 10K

18:39

sprint, and then I think we followed up

18:41

with another 15 grand. So, we had like

18:45

an initial 25K build, and this was

18:47

already like straight-up cash for us.

18:51

And then, with the follow-up clients,

18:53

what we were doing is we would purely

18:55

priced on tickets processed, knowing

18:59

that as people, as companies start to

19:01

process more tickets with AI, so more

19:03

tickets flow through our system, the

19:05

more value they get from our system. And

19:08

this is another software hack that's is

19:11

is always good. Whenever possible, you

19:13

always want to tie the pricing model to

19:15

the value. So, that as people use the

19:17

software more, as people get more value

19:19

from the software, the price goes up.

19:21

So, then if we zoom out and look at all

19:23

of this, step number two, your ideal

19:26

outcome should be three to five near

19:28

identical customers, and they should be

19:31

happy, they should be paying, and they

19:34

should give you testimonials. If you

19:37

have all this, then you are probably in

19:39

a better position than 99% of the people

19:41

that try to start a software company,

19:43

and you're in a great spot to continue

19:45

to really scale this thing. And this is

19:47

actually the perfect moment to talk

19:48

about today's video sponsor, Vanta,

19:50

which is a company I'm working with

19:51

right now. Because here's what almost no

19:53

one warned you about when you're trying

19:55

to sell to more companies, especially

19:57

bigger companies, is that once you have

19:58

that product market fit, they stop

20:00

asking, "Will this work?" and "How does

20:03

it work?" They start to ask like, "Is it

20:05

secure and can you prove it?" That's

20:07

when you suddenly need things like SOC 2

20:09

or ISO 27001, and whether you have them

20:12

can really make or break a deal. And

20:14

with AI baked into literally everything

20:16

right now, not only your own solution,

20:18

but also the services and tools you are

20:19

using to build and run the company, your

20:22

risk surface is expanding fast, and your

20:24

compliance needs to keep up with that.

20:26

And most people have no idea how that

20:27

works. And now, I'm going to be honest

20:29

with you, for Glide the software product

20:30

that I'm focusing on right now, none of

20:32

us on the team really started out as

20:34

compliance experts. So, that's really

20:36

why we started to use Vanta, because it

20:38

gives us one place to see our entire

20:40

security posture, meaning our internal

20:43

risks, our third-party vendor risks, and

20:45

continuous monitoring running in the

20:47

background, so this whole thing doesn't

20:49

become this once-a-year thing when you

20:51

need to prepare for an audit, but it

20:53

constantly runs in the background. It

20:55

can automate the path to frameworks like

20:56

SOC 2, ISO 27001, and HIPAA, and it also

21:00

covers the newer AI governance

21:01

frameworks everyone's starting to pay

21:03

attention to, like ISO 42001, the EU AI

21:06

Act, and the NIST AI Risk Management

21:09

Framework. Vanta plugs into your stack

21:11

with over 400 integrations and more than

21:13

16,000 companies running on it,

21:15

including ones you probably know like

21:17

Ramp and Cursor. And working with Vanta

21:19

has been a complete game-changer for us

21:21

working on Glido. And as you know, I

21:23

rarely do sponsorships on this channel

21:25

and only really promote companies that I

21:27

actually work with and would highly

21:29

recommend. So, if you're building a

21:30

software product and you want compliance

21:32

to be taken care of, then make sure to

21:33

check out the link in the description.

21:35

vanta.com/dave.

21:37

You can get a thousand dollars off if

21:39

you get started with them. So, know that

21:40

compliance will start to play a bigger

21:42

role as you start to scale. And that's

21:44

when we are ready for actually step

21:47

number three. And this is where we're

21:50

going to move into the final stage. And

21:52

this is what I call the done by them.

21:56

So, remember, right? It's you do it for

21:59

them, you do it with them together, and

22:00

this is where you move done by them,

22:02

where you actually turn the product into

22:05

a software product with self-onboarding,

22:08

self-checkout. Pretty much the whole

22:10

goal here is that you work towards a

22:12

solution is where your customer can get

22:16

results without you being in the

22:19

picture. So, while you are out there on

22:21

a beach somewhere, wherever you want to

22:24

be on a beach, like there should be

22:25

someone that can go to a website, can

22:27

create an account, can sign up, and then

22:30

ideally get results as quickly as

22:33

possible. So, this and only at this

22:35

stage

22:36

are you actually going to spend more

22:38

time on usually the engineering apart

22:42

engineering part around things like the

22:45

onboarding and multi-tenant support. And

22:49

this is things where this is also the

22:51

stage where things like your funnel and

22:54

plus website are going to be more

22:58

important because you need to draw

23:00

people in. It needs to make sense.

23:01

People need to understand it. They need

23:04

to be able to log in with an email, but

23:06

they also like the sign up with Google

23:08

button, right? So, all this complexity

23:11

grows. And trust me, there is like going

23:15

from a done with them setup where you

23:19

have 80% of the build ready and you do

23:22

the last part for them going to like

23:25

fully done by them, most people

23:26

underestimate this. Because

23:29

even with what you can do with coding

23:31

agents nowadays, yes, you can move

23:32

really fast, you can build the MVP, but

23:35

everything that comes around it, things

23:37

like

23:37

emails on the onboarding, password reset

23:40

flows, creating the accounts, magic

23:42

links, two-factor authentication, sign

23:45

up, follow-up emails, all this

23:47

complexity around it, this is where a

23:50

lot of people actually get stuck and

23:52

underestimate what it really takes to

23:55

run and scale a software company. So,

23:58

that's why this is step number three,

24:01

done by them. So, really what you should

24:03

optimize for in this stage, you're

24:05

you're already making money, you already

24:06

have a couple of customers, how fast can

24:09

you give them that result? Because

24:12

really, most products, probably say 90%

24:15

of products lose there. So, what happens

24:18

within like the first 5 minutes that you

24:21

onboard to that application? Can people

24:24

figure it out themselves and start to

24:26

get a quick win or value from it? How

24:29

easy is it? And most people will simply

24:31

not stick with it beyond the trial phase

24:33

or the first month and they will churn

24:35

and you lose them forever. So, the

24:36

faster you can get them this win, the

24:38

better. Now, then, the next big thing

24:41

that you want to focus on is

24:43

distribution. And this pretty much means

24:45

how are you going to get your product in

24:49

front of more people. Now, this is not

24:51

going to be a video on how to market

24:53

your software business. There are like

24:55

so many ways to go about this, but I'm

24:57

going to give you one tip. Most of

24:59

successful products, they don't just

25:02

rely on like paid ads because this is

25:05

like this this is the easy option,

25:06

right? But it's also very costly usually

25:09

in the beginning to get this profitable.

25:13

So, the best products have a built-in

25:16

distribution mechanisms like referrals,

25:20

sharing it with people, getting some

25:22

types of bonuses. If you look at the

25:24

early days of Facebook, for example,

25:26

this is how they grew big because a

25:27

social media platform is only as good as

25:30

the amount of people that are on there,

25:31

right? So, they incentivize to invite

25:34

people on there and there's this network

25:35

effect. So, that is one thing that you

25:39

can try and optimize for if you can. A

25:41

network effect, another recent great

25:43

example is school platform to host

25:46

community and courses. On school, you

25:48

have a 40%

25:50

lifetime

25:52

commission

25:54

on every referral or sale you make. So,

25:56

if people sign up using your link, right

25:59

now the plan was always kind of like 99

26:01

bucks a month, you would get lifetime

26:03

40% off of that. So, that's a built-in

26:06

incentive. Not only has it does it have

26:08

a network effect because you can invite

26:10

people in the community, there's also a

26:12

motivation for people to start promoting

26:14

it because this is actual actual money

26:17

if you build that up. So, that's the

26:19

goal of phase number three. You're going

26:20

to move from doing some things with them

26:23

to people being able to fully onboard

26:25

them themselves and then also getting

26:27

value quickly. That is what you should

26:29

optimize for. Then on top to grow, focus

26:32

on a distribution mechanisms. A network

26:35

effect, a built-in distribution

26:36

mechanism is always the best and then

26:38

you can expand to other forms of

26:40

marketing. So, that concludes the

26:42

surface to software ladder. You start

26:45

level one, done for them, level two,

26:47

done with them, and then level three,

26:49

done by them. Now, the one thing to keep

26:51

in mind is that you can jump off or stay

26:55

at any of these levels. Not everyone is

26:58

going to make it to level three. Let me

27:00

actually rephrase that. Almost no one is

27:03

going to make it to level three by

27:05

definition because it's really hard to

27:07

build a sustainable growing software

27:09

business. And now the great thing is

27:10

that even at level one and level two,

27:12

that's a great business and you can make

27:14

great money there. Now, you may not get

27:16

to the multi-million dollar exit, but

27:19

let's be real. The chances of that are

27:21

very low anyway. And by following the

27:23

exact process that I've explained over

27:25

here, you can greatly reduce the risk.

27:28

You don't need external funding. You can

27:30

actually get paid to prove your idea and

27:32

have fun in the process without

27:35

disappearing for months, not getting

27:37

paid, and then figuring out that it

27:38

wasn't worth it at all. And that's why I

27:41

love this model and will continue to use

27:43

it as well with my software development

27:45

company. This is the exact process that

27:48

we continue to do over and over again to

27:50

spot opportunities and to try to turn it

27:53

into a product. And with that, we have

27:54

come to the end of this video. Now, I

27:56

know that most of the people watching

27:58

this video right now will kind of like

27:59

click off and think like, "Ah, maybe

28:01

someday." But I know that a small

28:02

percentage of you watching really wants

28:05

to take this seriously. And in order to

28:07

get started with step one, doing it for

28:09

them, one of my favorite ways to really

28:11

do this is to first position yourself as

28:13

a freelancer and get started with that.

28:15

So, I have an entire video here on

28:17

YouTube on how I would start over as a

28:19

freelancer, which is something I've been

28:21

doing for years already. I also help

28:22

other developers and data professionals

28:24

to get started with that. So, if you

28:25

want to know how to find your first

28:27

project as a potential opportunity to

28:29

start working on this, make sure to

28:31

check out this video next. It's here on

28:33

YouTube completely for free and I will

28:35

share everything that you need to know.

Continue with YouTLDR

Analyze another video with Pro

Process a new video, search every timestamp, compare sources, and keep the result in your library.

Get Pro — $12/month30-day money-back guarantee

More transcripts

Explore other videos transcribed with YouTLDR.