“We’ll Fix It Later” – How Long Have You Been Saying That?

I help founders fix the systems that make everything take twice as long -clunky workflows, delivery bottlenecks, the duct-tape holding everything together.

Yet for 2 months, I avoided fixing my own for this exact reason.

The backstory:

I hate paying for SaaS subscriptions. But I still do it for essential tools.

I also self-host and maintain my own websites. Yeah, I’m a nerd.

Until 2 years ago, maintaining my WordPress sites was a nightmare:

• Minor upgrades: 1+ hours, constant anxiety something would break
• Major upgrades: Forget it. Too risky.
• Result: Outdated sites getting hacked regularly
• Cost: $300/year on Linode

I was literally avoiding necessary upgrade for 7 years because the process seemed so painful.

Sound familiar?

Most founders I talk to have similar technical debt somewhere in their business. Maybe it’s your websites. Maybe it’s your CRM. Maybe it’s those “temporary” spreadsheets running critical operations.

You know it’s costing you time and money. You know it needs fixing. But the pain of fixing it seems worse than living with it.

Here’s what I did:

I defined the desired change:

1️⃣ Easy maintenance (actually easy, not “technically possible”)
2️⃣ No-fear upgrades with quick rollback
3️⃣ Cut hosting costs, at least 2 times
4️⃣ Simple setup/removal for new apps
5️⃣ Proper security (finally!)

I knew Docker could help, but DevOps isn’t my thing. So I asked ChatGPT to teach me – just enough to solve my specific problem. 6 hours of research & scripting, 2 weeks of stabilization.

Results:

✅ Maintenance: ~0 effort
✅ Upgrades: No fear, no drama
✅ Costs: $300 → $86/year (71% reduction) – moved away from Linode
✅ New apps: Deploy in minutes
✅ Security: Reverse proxy + Cloudflare = hackers went South

2 years ago, I would’ve paid someone $500+ to do this. Or actually… not done it at all.

So postponing the upgrade costed me around $2100 for those 7 years but could have costing around $600.

Last weekend reality check:

Upgraded n8n from 1.120.3 to 2.11.4 (including Postgres database migration 12→18). I’d been avoiding it for 2 months. The required database migration was exactly the reason I was postponing. So I spoke to Mr. Claude this time and it helped me build the migration plan and debug the migration issues. Tested. Done.

Total time spent: 2.5 hours. Feels great! 💗

The point:

This is what I help founders do with their operations – not necessarily the tech stack (unless you want that), but the systems that you’re avoiding because they seem too hard to fix.

• The deliveries that are “good enough” but fragile.
• The processes everyone knows but nobody documented (awaiting for the bus?).
• The operational debt that’s silently draining time, money, energy.

I bet I’m not the only one. 👀

So here’s my question:

What’s one system in your business that “just works”… but you know is held together with duct tape and costs you 3x of what it could?