Running your week: daily, weekly and monthly habits

A system is only as true as the habits around it. This is the rhythm that keeps the numbers worth reading — what happens daily, weekly and monthly, and who owns each part. Adapt the roles to your shop; the cadence is the part that matters.

Record movements immediately — everyone. Goods in when the van leaves, material issued when it is taken, output when the job is done, despatch when it ships. Not at the end of the day from memory. This is the whole foundation, and it is the one habit that cannot be delegated to a weekly catch-up.

Clear the shortage list — whoever owns purchasing. A short job found today is a phone call. Found on the morning of the build, it is a late delivery. Five minutes on this is the highest-value five minutes in the day.

Log issues where they happened — the shop floor. A scrapped part, a wrong dimension, a tool that failed. Logged against the job while the detail is fresh, this becomes evidence. Remembered later, it becomes an argument.

Run planning and look at what it recommends — production lead. The point is not to accept every suggestion. It is to see the gap between what is promised and what is possible while there is still a week to act on it.

Check capacity against the order book. Not "are we busy" but "is any work centre committed beyond what it can do". Overload found a week out gets solved with overtime or a conversation. Found on the day, it gets solved by disappointing somebody.

Count a handful of locations — warehouse. Not a full stock take. A rolling handful, chosen from your fastest-moving parts. This is how you find out that discipline has slipped, four weeks before it would have shown up as a mystery shortage.

Invoice what shipped. A week is the longest this should ever wait. It is also the cheapest cash-flow improvement available to most small manufacturers.

Look at job costs against quotes — owner or manager. Which jobs made money and which did not. This is the report that changes what you quote, and the one most likely to be skipped because nothing breaks if you do.

Review what got scrapped and why. One month of issues is noise; three months is a pattern. Patterns point at a supplier, a machine, a drawing or a training gap.

Chase receivables properly. Work through what is overdue rather than reacting to whoever shouts.

Fix the BOMs the month proved wrong. Every month of real building surfaces recipes that are subtly off. Correcting them as you go is the difference between a system that gets more accurate over time and one that decays.

Every habit above needs a name against it. "The team" owning stock accuracy means nobody owns it. The specific failure mode is quiet: nothing breaks for weeks, then availability is wrong, and by then the cause is a hundred small unrecorded movements that cannot be reconstructed.

If you are small enough that one person owns most of this, write it down anyway. The list is what you hand over when you hire.

What's next