THE PULSE
1. Most small businesses are running on stale numbers and don’t know it.
A 2026 bookkeeping benchmark study of professional services firms found only 8% reconcile their accounts daily and 24% weekly. 42% do it monthly, and 7% reconcile quarterly or less, a group the report itself flags as “operating with dangerously stale data.” Source: Steph’s Books, 2026 Bookkeeping Benchmarks.
What it means for you: If you’re in the monthly camp, and most owners are, the number you’re making calls on today is already three to six weeks out of date by the time you see it.
2. How fast you close your books predicts how fast you grow.
The same study found the median monthly close time for firms doing $1M to $10M in revenue is 14.2 business days. Top-quartile firms close in 6 days or fewer, bottom-quartile firms take 22 or more. Firms closing within 7 days were 2.3 times more likely to report revenue growth above their industry median. Source: Steph’s Books, 2026 Bookkeeping Benchmarks.
What it means for you: This isn’t a bookkeeping nicety, it’s a growth lever. The businesses seeing their numbers fastest are also the ones growing faster, and that’s not a coincidence. It’s the sequence.
3. A small entry error is cheap. A small entry error nobody catches is not.
Manual data entry runs a 1 to 4% error rate at the field level under real-world conditions. Catching a mistake at the point of entry costs $1 to $5 to fix. By the time that same error reaches a report, a filing, or a decision, it costs $50 to $500 or more to unwind. Source: Lido, July 8, 2026.
What it means for you: Every unreviewed line in your books is a small bet sitting open, and the cost of that bet multiplies the longer it goes unnoticed. That’s exactly what happens when nobody’s looking.
YOUR DASHBOARD IS LYING TO YOU, CALMLY AND CONFIDENTLY
You didn’t make a bad decision last quarter. You made a good decision on a bad number. Those cost you the same amount, but they are not the same mistake, and only one of them you can actually prevent going forward.
Pilots fly on instruments they trust completely, and that trust is earned the hard way. Every instrument gets calibrated on a schedule, logged, and checked, because nobody’s allowed to just assume the altimeter is still telling the truth. Most owners run their whole business on one instrument, the number sitting in their accounting software, and never once ask when it was last calibrated.
I had a client in a wellness services business who’d priced her signature service by marking up the product alone. No overhead, no labor cost for the person performing the service, no sales tax factored in anywhere. On paper the price looked fine, competitive even. But she didn’t have the buying leverage her bigger competitors had, so her actual cost to deliver was significantly higher than theirs, and matching their price meant she was roughly breaking even on a service she believed was one of her better earners. The number never announced any of that. It just sat there looking fine, month after month.
Once we saw it clearly, there were really only three moves on the table: go upstream and find a cheaper way to source the product to protect the margin, bundle the service with something else so the blended margin actually worked, or stop offering it. That one mispriced service was quietly costing her thousands of dollars a month, and it was most of the reason her cash always felt tight by the end of the month.
The report doesn’t announce that it’s wrong. That’s what makes it dangerous. A wrong number looks exactly like a right number: same font, same clean columns, same reassuring total at the bottom. Nobody opens their books expecting to be lied to, so nobody double-checks, so the lie just sits there, collecting interest.
THE VERDICT: a real number, graded.
The decision: In June, AeroVironment disclosed it had understated its net loss by roughly $87 million after excluding a piece of goodwill from an impairment calculation tied to its Space division, following the termination of a Space Force contract. The company restated multiple quarters, and the stock dropped 11% the day it came out.
My verdict: The market didn’t punish AeroVironment for losing the contract. It punished them for finding out, four quarters later, that they never actually knew how much they’d lost.
Why most people are reading it wrong: The headline is “accounting error.” The real story is “material weakness in internal controls,” company language for we didn’t have a reliable process checking this number before it went out the door. This is a company with a finance department and outside auditors, and the number was still wrong for the better part of a year.
Your version: You don’t have an auditor combing your books line by line. You have you, glancing at a dashboard between calls, trusting it because it’s never occurred to you not to. If a team of professional accountants can miss $87 million for a year, the number you’re eyeballing on your phone at a stoplight has earned some healthy suspicion too.
How to check if your own numbers can be trusted
You don’t need to become a bookkeeper for this. You need forty-five minutes and the willingness to ask your own numbers an uncomfortable question: prove it.
1. Find the last reconciliation date. Open your accounting software and check when your bank and card accounts were actually reconciled, not just “connected.” Past a month, everything downstream of that number is a guess wearing a costume.
2. Pick the one number you check most and trace it to its source. Cash balance, gross margin, whatever you glance at weekly. Follow it back to the real transactions behind it. You’re not hunting for fraud, you’re hunting for lag, miscategorization, and duplicates, the boring stuff that quietly compounds.
3. Spot-check three transactions at random. Confirm each is categorized correctly and matches a real receipt or invoice. If even one is wrong, assume the same error rate applies to everything you haven’t checked.
4. Ask when the number was calibrated, not just updated. A number can update every day and still be wrong every day if nobody’s checking the inputs behind it.
5. Put a standing recheck on the calendar. Weekly beats monthly, monthly beats “whenever tax season reminds me.” The businesses closing fastest aren’t smarter. They’re just not letting the gap get wide enough to hide something in it.
THIS WEEK’S MOVE
Pick the one number you rely on most to run this business, cash balance, margin, pipeline value, whatever it is, and trace it back to its source today. If it holds up, you’ve bought real confidence. If it doesn’t, you just caught something before it became a decision you couldn’t take back.
Best,
Micah
THE BROWN BOX
The Brown Box is out now, and “Uncover Numbers” is step two of the TURNAROUND method for exactly this reason. You can’t fix what you can’t see clearly, and most owners are flying on instruments they’ve never once recalibrated. If this issue hit close, that chapter goes further than I could here.
Grab your copy → https://www.amazon.com/dp/B0H2X4Z85W
Already read it? A two-line review does more for this book than almost anything else you could do for me. Hit reply once it’s up, I read every one.
