The 90-Day Build: SmallBizOps

February 8, 2026
Day 0 of 90

For 15 years, I’ve solved operational chaos.

Not from a whiteboard in a conference room. From the factory floor. The warehouse. The back office where “the system” breaks down and people improvise with spreadsheets, sticky notes, and hope.

I’ve built ERP and CRM systems for food manufacturers navigating FDA compliance, automotive shops juggling estimates and parts inventory, language schools coordinating international students across time zones.

The pattern is always the same: smart people trapped by inadequate tools.

They don’t need “enterprise software.” They need systems that match how they actually work.

So I build those systems.

The Challenge

Here’s the problem with custom business systems: nobody sees them.

I’ve architected platforms that reduced manual rework by 30%. Built dashboards that turned hours of reporting into real-time visibility. Designed workflows that eliminated entire categories of operational errors.

But it all happened behind closed doors. Private companies. Proprietary systems. NDAs.

You have to take my word for it.

That ends now.

The Plan

For the next 90 days, I’m building SmallBizOps—a lightweight operations toolkit for small businesses—completely in public.

Not a prototype.
Not a proof-of-concept.
A production-ready system, built the way I’ve built systems for 15 years: solve the problem first, write the code second.

Every decision documented.
Every feature shipped on schedule.
Every line of code visible on GitHub.

What I’m Building

SmallBizOps is for the business that’s outgrown Excel but can’t justify enterprise software.

The shop owner tracking inventory in three different spreadsheets.
The manufacturer running operations through email and hope.
The service business where “customer database” means scrolling through Gmail.

I’ve seen this exact chaos in a dozen industries. The problems are universal. The solutions shouldn’t require six-figure budgets.

The System (12 weeks, 4 core modules):

Weeks 1-3: Inventory Management

• Product catalog with categories and SKUs (done)
• Stock tracking with intelligent low-stock alerts (done)
• CSV import (because everyone’s data lives in Excel first) (done)
• Product images and specifications (done)
• Search, filter, bulk operations (done)

Why this first: Inventory is the foundation. Get this right, everything else gets easier.

Weeks 4-6: CRM Foundation

• Customer database (contact details, purchase history, notes)
• Communication timeline (every interaction, timestamped)
• Segmentation and tagging (“wholesale,” “VIP,” “seasonal”)
• Email templates for common scenarios
• Basic pipeline visibility

Why this matters: CRM isn’t about “managing relationships.” It’s about institutional memory. When Sarah leaves, the new person knows what happened with the Johnson account.

Weeks 7-9: Invoicing & Revenue

• Invoice creation (pull customers + products, generate instantly)
• Professional PDF output
• Payment tracking (sent → paid → overdue)
• Revenue analysis (by month, customer, product line)
• Email delivery from system

Why this works: Invoicing isn’t accounting. It’s operations. Faster invoicing = faster payment = better cash flow.

Weeks 10-12: Executive Dashboard

• KPI cards (customers, inventory value, monthly revenue)
• Trend charts (what’s moving, what’s not)
• Export-to-Excel for deeper analysis
• Customizable date ranges
• Operational health at a glance

Why dashboards last: You can’t visualize what you haven’t captured. Data first, insights second.

The Approach: Process Before Code

Here’s what 15 years taught me:

Bad approach:
“Let’s build a CRM. What fields should the customer table have?”

Right approach:
“Walk me through what happens when a customer calls with a complaint. Who needs to know? What gets recorded? Where does the information go next?”

Map the process. Find the pain points. Then write code.

This is how I’ve always worked:

1. Understand the actual workflow (not the “official” one)
2. Identify what breaks, what’s manual, what gets forgotten
3. Design the system around how people actually work
4. Build only what’s necessary
5. Train the team, iterate based on reality

SmallBizOps will be built the same way—except this time, you see the entire process.

The Tech Stack

I’m not chasing trends. I’m using what works in production:

• Laravel 11 (PHP 8.2) — mature, well-documented, battle-tested
• MySQL 8 — relational data for relational problems
• Blade + Tailwind CSS — fast UI, no JavaScript framework complexity
• Railway.app — simple deployment, zero DevOps overhead

No microservices. No “we’ll scale later.” No resume-driven development.

Just clean, maintainable systems that solve real problems.

The Rules

1. Public by default

• GitHub repository: github.com/drfb02/smallbizops
• Every commit visible
• Every architectural decision documented
• Weekly progress reports (every Saturday)

2. Ship on schedule

• Week 3: Inventory module live and usable
• Week 6: CRM module live and usable
• Week 9: Invoicing module live and usable
• Week 12: Complete v1.0 release

Done beats perfect. Shipped beats planned.

3. Production quality

• Code I’d deploy for a paying client
• Proper validation, error handling, security
• Documentation that explains why, not just how
• No “TODO: fix this later” in production code

4. Real-world testing

• By Week 6, at least one small business using it for real operations
• Iterate based on actual usage, not assumptions
• Features driven by genuine need, not “nice to have”

Why This Matters

For small businesses:
You get a free, open-source tool that solves real operational problems. Self-hosted, customizable, no monthly fees.

For the industry:
A public case study of how to build business systems the right way—process first, code second.

For me:
Proof of concept. Not “I can code Laravel”—that’s table stakes. Proof that I can architect operational systems, ship on deadline, and solve messy real-world problems.

After 15 years of building in private, it’s time to build where people can see the work.

The Timeline

February 9, 2026 — Day 1
Database architecture, schema design, first migrations

February 15, 2026 — Week 1 Complete
Core inventory CRUD operational, UI foundation established

March 1, 2026 — Week 3 Milestone
Inventory module production-ready. Deployable, documented, testable.

March 22, 2026 — Week 6 Milestone
CRM module production-ready. Customer management fully operational.

April 12, 2026 — Week 9 Milestone
Invoicing module production-ready. End-to-end revenue workflow complete.

May 3, 2026 — Week 12 FINAL
SmallBizOps v1.0 released. All modules integrated, documented, deployed.

Success Metrics

By Day 90, I’ll have:

✅ A complete, working system (not a demo)
✅ 200+ commits with clear documentation
✅ Live demo environment anyone can test
✅ Deployment documentation for self-hosting
✅ At least one small business running real operations on it

The real measure: Can someone fork this repo, deploy it, and run their business on it without calling me?

If yes, it worked.

Follow the Build

Code: github.com/drfb02/smallbizops (daily commits)
Blog: Weekly updates every Saturday (technical deep dives, lessons learned, architectural decisions)
LinkedIn: francesco-boschi-bb5937358 (weekly progress, Saturday recaps)

Day 1 Starts Tomorrow

February 9, 2026, 10:30 AM.

First commit: database schema for products and categories.

Twelve weeks. Four modules. Zero excuses.

Let’s build something real.

— Francesco Boschi
Business Systems Engineer
15 years solving operational chaos
Now building in public

P.S. — If you run a small business and this sounds useful, star the repo. I’ll prioritize features based on real operational needs, not theoretical nice-to-haves.

P.P.S. — If you’re technical and want to follow the architecture decisions, watch the repo. I’ll be documenting not just what I build, but why I make specific design choices.

Next post: February 15, 2026 — “Week 1: Database Design for Real Operations”