Will We Last the Year?
Two entities · ~$2.5M revenue · three-month diagnostic.
The owner had one question: would they still be in business in twelve months?
Their own books couldn't tell them. Margins swung from 30% to 70% month over month. The income statement looked like a different business every period — not because the data was missing, but because nothing was being recorded when it actually happened.
Numbers reached the owner twenty to thirty days after month end. By then the month they described was over and the decisions were already made. That came down to about ten days, which is the change everything else here depended on — a forecast built on figures that show up a month late is a history lesson.
What turned up
Cost of goods sold was being booked the day inventory arrived, not the day it sold. A heavy buying month gutted the P&L. A quiet buying month made the business look like a software company. The owner had been making decisions off numbers that described stockroom activity, not retail activity.
The POS system held the real story — by transaction, by hour, by SKU. None of it was reaching the accounting books in usable form. Revenue showed up. Margin didn't.
What got built
- A twelve-month trailing trend analysis to expose the real rhythm underneath the noisy margins
- A six-month forward revenue forecast tied to historical seasonality and current trajectory
- A rebuilt COGS workflow that recognized cost when product sold, not when it arrived
- POS data flowing into accrual revenue and COGS on a repeatable cadence
- Customer traffic mapped by day-of-week and hour-of-day, feeding directly into staffing decisions
- Their CPA re-run on the corrected numbers, so the tax position matched the books rather than the old ones
What they walked away with
An income statement they could read. A six-month forward view of revenue to plan inventory and staffing against. Staffing decisions backed by their own customer data instead of guesses.
And the answer to the question they came in with — built on numbers that finally meant what they appeared to mean.
Worth being straight about where this one ends. The engagement ran three months, the rebuilt COGS workflow and the forecast were handed over, and it finished there. How the margins settled afterward is genuinely unknown — nobody was there to see it. What can be said is what got built and what it did to the close. Anything past that would be taking credit for someone else's year.