The Operations System Behind a Bootstrapped 11-Site, ~$4M Business

A two-warehouse ERP model that took stock variance from 4–7% to ~0%, cut reporting lag from 17 days to 1, and reduced related payroll ~14%

6 min read
Published September 8, 2026
operations systemsERPinventory managementunit economicsbootstrappedmulti-location operations

TL;DR

As Co-Founder & Operations Lead of Taboo Bar Group (Feb 2018 – May 2024) I scaled 1 → 11 locations to ~$4M annual revenue with no outside capital — nine sites in the first two years. The core enabler was an operations system: a two-warehouse inventory model inside the ERP that took draught stock variance from 4–7% to ~0%, reporting lag from 17 days to 1 day, and related payroll down ~14% — with no new software purchased and no headcount added.

The problem

Multi-site hospitality lives or dies on stock control. With 15+ taps per location, the standard industry answer is periodic manual stocktakes: slow, late and wrong. We were seeing 4–7% variance between what the system said and what the warehouse held, and operating data arrived 17 days after the fact — far too late to act on.

The system

The fix was structural, not heroic: each location was modelled as two virtual warehouses in the ERP — storage and bar — with every internal movement recorded as a transfer. That single modelling decision made variance visible per stage instead of per month, and turned the stocktake from an investigation into a reconciliation.

  • Draught stock variance: 4–7% → ~0% across all sites.
  • Operating-data lag: 17 days → 1 day.
  • Related payroll: down ~14% (roughly 237 staff-hours a month removed).
  • Supplier-financed inventory: a 7-Eleven-inspired replenishment model with ~3-day turnover kept cash in the business — each new site effectively self-funded.

Scaling it

The system was the repeatable part: full P&L and per-location unit economics, a remote-opening playbook, and an organisation built to 50 people across 1,700+ interviews. It ran well enough remotely that new locations opened while I operated from another country.

Why this matters for a startup

Capital efficiency is a system property, not a personality trait. The same pattern — instrument first, make data real-time, remove manual reconciliation — is what a modern GTM stack needs: a single source of truth, enrichment that stays clean, and reporting the founder actually reads. That's the discipline GTM Stacker sells.

The full write-ups of this system — the inventory model, the direct-costing approach and the cash-flow lessons — are on the Writing page.
Share:
About the Author

Theo Popov runs GTM Stacker, a growth-systems practice for startups and venture programs. Previously Co-Founder & Operations Lead of an 11-location, ~$4M hospitality group bootstrapped with no outside capital, he now builds AI-native enrichment, outbound, content and reporting systems. More about Theo →