The Audit Room and the Kitchen
What statutory audit taught me about the gap between a business's story and its numbers.
The high floors of a bank at midnight during audit season are quiet in a specific way — everyone in the room is doing arithmetic, and nobody’s going home until it agrees.
That’s where I learned there are two ways to check if a number is true. Test the controls — check the machine that produced it, and trust the output if the machine is sound. Or test substantively — ignore the machine, go to the source records, rebuild the number yourself. Controls testing is cheap. Substantive testing is certain. Every audit is really an argument about how much you’re willing to believe a system’s account of itself.
I left in 2009 and spent every year since inside businesses that run on narrative — hospitality is a narrative business. I never stopped running the same two tests.
Opening a store is substantive testing. I opened the first three Pita Pit stores in India myself. You count the covers. You watch the checklist get completed because you’re the one completing it. Nothing is reported to you — you are the record.
Then I went international, and it inverted. Regional director across India, the UK, the UAE, Saudi, Singapore — I couldn’t stand in those rooms anymore. What I had instead was the playbook: SOPs, training, reporting cadence. I stopped verifying outcomes and started verifying the machine that produced them. That’s controls testing, and it’s the only thing that works at distance.
But trusting systems for years does something to you that doesn’t announce itself. By the time I was signing franchise agreements, I trusted the paperwork too. I ran the substantive test on the thing I knew how to test — unit economics, cover counts, payback period — and the numbers held. They held because they were the wrong numbers. A family investor with capital isn’t an operator, and nothing in a diligence pack tells you which one you’re signing. That cost us stores. It was a controls question wearing a substantive question’s clothes, and I tested the wrong one.
Later, building cloud kitchens, I learned the same lesson from the other side: 122 kitchens, three cities, ninety per cent occupancy inside a year — but occupancy only means something once you know what it’s a ratio of. A number without its denominator is a story wearing a number’s clothes.
Capital and execution speak different languages — one in multiples and cohorts, the other in covers and roster hours — and most businesses don’t fail at strategy or operations. They fail in the translation between them. I’ve spent two decades moving between those two rooms. That translation is real work, it’s nobody’s job title, and it’s where most of the value goes missing. That’s the gap I build in.