It is the last day of September, and this morning I wanted a straight answer to three questions: what shipped this month, what slipped, and what changes in October.
Without a record, that is a week of work: scrolling through chats, opening three apps, asking people what they remember, and still missing half of it.
This morning it took minutes, and not because I have a better memory. Every day this month wrote one entry into one place: what went out, what was skipped and why, what went wrong. Twenty-nine entries before today's, one per day, no gaps. The month-end review was not an exercise in remembering. It was reading.
What shipped
Two pieces of writing a day, every day, and the short videos and images that go with them. I will not repeat the counts here; the point of this post is where they came from, not how big they are.
The bigger thing that shipped was a decision. On 25 September I narrowed everything we publish to the two things we actually sell: NexBrand, one inbox for every way a customer reaches you, and the Business Operating System, one system for how the business runs. I wrote about why in Eighty-nine posts and not one of them named what I sell.
What slipped
The log is honest, which is the only reason it is useful.
- A video went out with speech on it. Gibberish, in what sounded like another language. A check had said it was clean. I wrote that one up in The check said clean. It was asking the wrong question. Every video now ships silent, with the words on screen.
- The plan was written for a business we had stopped being. September's calendar was drafted before the 25th, so for the last five days almost every planned topic had to be swapped for one about the two products. Each swap is in the log with a reason. Today's was one of them.
- One day started in the afternoon. Two morning slots had already passed, so they moved later. Nothing was lost, and the log says exactly which slots moved and why.
None of these was a disaster. Each would have been invisible, and repeated, if it had lived in someone's head instead of on the record.
What changes
October was planned the other way round. It starts from the two products and works back to the topics, instead of starting from topics and hoping they point somewhere useful. It was approved before the month began, not patched as it ran.
Why I am telling you this
In a lot of businesses, those three questions take a week of digging. Not because anyone is disorganised, but because the answers are spread across a phone, an inbox, a spreadsheet and the memory of whoever was there.
That is the whole case for one source of truth. When every piece of work leaves its record in one place as it happens, the month-end review stops being an investigation. It becomes a read. That is what a Business Operating System is for, and it is what I get to use on my own business every day.
Frequently Asked Questions
What should a month-end review answer?
Three things: what was delivered, what slipped and why, and what changes next month. If answering them means asking people to remember, the records are in the wrong places, not the people.
Why keep a log of what went wrong, not only what shipped?
Because slips repeat when nobody writes them down. A short note with the reason turns a bad day into a rule for next month, and makes it visible before it becomes a pattern.
How does one source of truth help at month end?
When every job, message and invoice leaves its record in one system as it happens, the month-end review is reading, not reconstructing. Nobody has to chase three apps and two memories for the answer.