48 Hours to Replace Salesforce

48 hours.

That's how long it took my Chief of Staff and I to vibe-code a CRM that — for our unique needs — is 10x better than Salesforce. It cost us about $100.

We didn't plan to build a custom CRM. What started as a simple Wednesday-evening request to visualize DoW org structure had become a functioning CRM by Friday afternoon. Another week of sporadic polishing in our free time and Waypoint — our defense-tech optimized CRM — was done.

We were genuinely blown away. In the year prior we'd taken calls with Salesforce, HubSpot, and others. Thousands of dollars. Lengthy customization processes. Third-party setup consultants. And the deep feeling that after all that effort we'd still be left forcing the way we do business into a generic CRM... staring and wishing we could change the features we didn't like. So we kept using Airtable in the interim — until our bills started ballooning into hundreds of dollars a month despite active users on our team being <5 people.

What we accidently built instead was self-hosted, essentially free, modifiable in real-time, and built around OUR business process, informed by our team's experts, and how DoW orgs and government procurement actually work.

For about 30 seconds we considered monetizing Waypoint. Then we remembered we built it in a week. 🫠

So instead I built a quick webpage (also vibe coded..) to share the source code that might benefit other founders and defense-tech startups annoyed by expensive, one-size-fits-none tools. Demo, fork, customize, deploy. Called it vibecache.

Waypoint is the first project there. Several others are already in the works.

Not sure where this goes from here, but it felt worth sharing. The experience convinced us this shouldn't be a one-off. We're now asking ourselves the question: "How much of our paid SaaS stack can we simultaneously eliminate and improve? Accounting, FP&A, ERP, engineering test, data analysis, etc?"

Have others had a similar experience?

Still weighing the macro ramifications of this, but it certainly doesn't look promising for traditional SaaS... 🪦