A few things the post doesn't cover that might be useful context:
What we got wrong with V1 and V2. We built a browser automation tool that depended entirely on platforms we didn't control. That's an obvious mistake in hindsight, but at the time, every company in our space was doing the same thing. The C&Ds were inevitable. We just didn't want to see it.
Why we chose a spreadsheet interface for V3. We tested three approaches: a workflow builder (what V2 was), a dashboard UI, and a spreadsheet. The spreadsheet won because our users already think in rows and columns. Every agency we talked to was exporting CSVs from 4 different tools and joining them manually in Google Sheets. We realized the spreadsheet IS the workflow, not a step in one.
*Where we are financially*. 25 people, bootstrapped, no VC. We've been close to zero runway multiple times. V3 is live with paying customers but we're early. I won't pretend otherwise.
*What I'd love HN's input on*. We're building waterfall enrichment across multiple data providers with stop-on-success logic and AI columns that run LLM calls per row. If anyone here has dealt with similar orchestration problems at scale (provider failover, cost optimization across APIs, batching LLM calls), I'd genuinely love to hear how you approached it.
Ask me anything about the legal battles, the tech, building B2B from India without venture money, or where we screwed up.