Negócios

I Didn't Stop Writing. I Was Building.

23 de fevereiro de 2026
7 min de leitura
saasarquitetura_de_softwarereacttecnologia

It's been about two weeks since the blog has been silent.

No new insights. No published reflections. No technical provocations.

It wasn't a lack of topics.

It was focus.

I was building something that had been maturing for months: STRATA.

The problem that took me away from the blog

For years, I've seen the same scene repeat itself:

Freelancers, small agencies, and extremely competent digital professionals... but operating in an improvised way.

Proposals made in Google Docs.

Versions scattered in folders.

Manual follow-ups.

No traceability.

No real conversion metrics.

The work is good.

The operation doesn't keep up.

A fragile operation limits growth.

That's where STRATA was born.

What is STRATA, really?

STRATA is not just a proposal generator.

It's an operational platform for those who sell digital services.

You create structured proposals, send them directly to the client, track real-time metrics (views, interactions, acceptance), and centralize the commercial flow.

But the central idea isn't "creating pretty PDFs."

It's allowing a freelancer to operate with the organization of a multinational.

Process.

Clarity.

Metric.

Standardization.

Without losing agility.

Building like an engineer, not a launcher

If you follow my insights, you know I don't take superficial decisions.

And that became even more evident while building STRATA.

It wasn't "let's upload something quickly and see what happens."

It was: let's structure it right from the start.

Not just to function.

But to sustain.

Main stack

The base was built with:

  • Next.js 16 (App Router)

  • React 19

  • TypeScript

  • Tailwind CSS

  • shadcn/ui + Radix

  • Prisma ORM

  • PostgreSQL

  • Vercel

  • Git/GitHub

But stack, by itself, doesn't say much.

What matters are the criteria behind it.

Engineering oriented to predictability

From the start, the priority was to reduce ambiguity and increase control.

  • TypeScript in strict mode

  • Zod for structural validation

  • @t3-oss/env for secure environment variable typing

  • server-only to prevent sensitive logic leakage

  • NextAuth + Prisma Adapter for robust authentication

  • Postgres as a solid relational database

Less improvisation.

More explicit contract.

Observability and monitoring from the first deploy

I didn't want to discover problems through users.

So:

  • Sentry for real-time error tracking

  • Uptime Robot with hooks that automatically trigger failures

  • Configured alerts for immediate analysis

  • PostHog for precise behavioral analysis

This completely changes how you evolve the product.

You don't react.

You anticipate.

Some important decisions

1. Server-first whenever possible

Architecture prioritizing Server Components and Server Actions.

  • less JavaScript on the client

  • less attack surface

  • less state inconsistency

  • better performance

Security and predictability go hand in hand.

2. Performance as a premise, not as a final adjustment

  • Priority Hints

  • Strategic lazy loading

  • Font optimization

  • Conscious animation control

  • Real code splitting

Commercial proposal cannot freeze.

Performance here is not a luxury.

It's credibility.

3. Security as architecture, not as a plugin

Along with structured authentication:

  • HSTS enabled

  • Configured headers

  • Server-side validation

  • Rigid access control by session

  • Prisma avoiding improvised queries

Minimum surface.

Maximum predictability.

4. Operation before feature

I resisted the temptation to add "interesting" features.

I prioritized:

  • clear flow

  • reliable generation

  • stable sending

  • real-time tracking

Feature that doesn't improve operation was left out.

Because product doesn't grow with brilliance.

Grows with structure.

What building a SaaS reinforced in me

Developing STRATA wasn't just about code.

It was about:

  • prioritization

  • real trade-offs

  • decisions under uncertainty

  • simplifying when the desire was to sophisticate

It was a practical laboratory for everything I've been writing about technical decision-making.

Building a product of your own changes your perspective.

You feel the weight of choices.

You understand the cost of rework.

You learn to say "no" to yourself.

Why I'm sharing this?

Because I don't want my insights to be just theory.

STRATA is the materialization of what I believe about:

  • clarity

  • structure

  • decision

  • technical responsibility

If you're a freelancer, small agency, or digital professional and feel that your operation could be more organized, it might be worth testing.

Invitation

If you follow my texts here, I'd like to invite you to test STRATA firsthand.

Access:

👉 https://www.getstrata.com.br

Explore.

Create a proposal.

Test the flow.

And, if you want, send me direct feedback.

Because building is good.

But evolving with real feedback is even better.

0

Compartilhar em:

Publicado em 23 de fevereiro de 2026
7 minutos de leitura

No spam. Only content worth opening.

Continue lendo

I Didn't Stop Writing. I Was Building. | Jhonatan Oliveira