7 steps to building a business

Do not index

Anyone remember how building a startup felt like back in 2021?

No ChatGPT. No Claude.

Just you, your laptop, and the blinking cursor on VS code.

Simpler times really.

That isn’t the case in 2025:

10s of tools. Multiple ideas. A lot of confusion.

Everyone asks me how to start a business in 2025.

So here's my exact 7 step playbook 👇

Step 1: Know your edge

If I had to define business success in 1 phrase:

“You have to be better than others at something.”

Start by understanding what makes YOU unique.

And then we’ll double down.

What skills do you have that others don't?

Do you have an audience already?

Any special market knowledge?

A network you can leverage?

A unique combinaison of skills?

Find your unfair advantage. That's your starting point.

Step 2: Build your idea bank

2025 is ruled by idea guys.

Building is not a bottleneck any more.

Knowing what to build is.

You need to collect new (and more) ideas, and they are everywhere if you know where to look:

Reddit questions (goldmine of problems)

Feature requests on Shopify, Webflow, etc.

Chrome extension store (what could be standalone?)

Amazon reviews (look at 3-star reviews of popular products)

App store complaints

Watch what people complain about.

Their ignorance is your opportunity.

Step 3: Quick validation

Okay slow down now, coder! Everything you think you don’t have to code.

What's the use of building if no one is buying?

We have to gauge the demand for the product before we actually start building it .

Here’s how you do that:

Use V0 or Lovable to create a landing page

Add one email input + "Request Access" button

Share in DMs, tweets, Reddit, etc.

See who gives you their email

If people drop their email = validation. No interest? Skip to next idea.

Because if people are not willing to commit to your product for free, how do you expect them to pay at once when you finally release the product.

Step 4: The One-Week build challenge

Pick the landing page that collected the most emails.

Then set a constraint: "What can I build in ONE WEEK?"

Not 2 weeks. Not a month. Not a quarter.

Just. One. Week.

Why!? 👉 It will force your to narrow it down to its core value proposition.

Build it. Ship it. Get it in front of real people.

Yes, it will be crappy. But you'll learn so much.

Step 5: Find the REAL problem

Most founders miss this:

The problem you think you're solving usually isn't the real one.

In the end, most problems come down to:

Money

Health

Sex (relationships/status)

Which one are you actually solving?

Go deeper. Iterate. Collect feedback. Repeat.

Step 6: Expand acquisition

When you find your core reason to buy and positioning — time to turbocharge your client acquisition.

Speed is the most important.

Because categories get saturated and sophisticated before you know. You might as well be the leader when it happens.

Once you understand your direction:

Build in public on X

Start with basic Meta & Google ads

Activate SEO (use Feather.so / Outrank.so to start quickly)

Create mini-tools for lead generation

Work with influencers and setup an affiliate program (use Rewardful)

Build virality into your product (make it sharable)

Remember: Each channel properly built compounds.

Step 7: Fix Retention

At this point, you'll have traffic and maybe (hopefully!) revenue.

But your retention will likely suck.

This is where most founders fail - they keep acquiring users who just leave.

Work obsessively on retention. It's your oxygen.

That is how your business compounds.

That is how you get top live that ‘digital nomad’ life.

That is how you get financially free.

Retention > acquisition.

Most people quit right before things get good.

The difference between success and failure isn't usually the idea.

It's consistency. Persistence. Adjustment.

PS Just curious - do you use AI to code, or is it just a gimmick for you still? Reply to this email and tell me 👇🏻

Tweet of the week

The world is changing faster than we know.

What side of history are you choosing?

Keep building

Tibo 💻

P.S. This probably looks simple. It is. But simple ≠ easy.