The Week After A Good Month

The month was excellent. The following one is disappointing, and nobody can say why.

Part of it is arithmetic. Part of it is a set of decisions that get made when things are going well, and the second part is worth separating out because it is the part you control.

The arithmetic part

An exceptional month is exceptional partly because several things went right at once. The next month has ordinary luck, and ordinary looks like decline after exceptional.

⚠️ This is not a problem and does not need fixing. It needs recognising, because treating normal variance as a failure produces worse decisions than the variance ever did.

The test: compare against the same month last year, not against last month. If this month beats its own equivalent, nothing has gone wrong.

The decisions part

This is the part that matters, and it happens quietly during the good month.

The pipeline stops being fed. Marketing, outreach, follow-up — all feel unnecessary when the room is full. ⚠️ They have a lag, so the gap appears six to eight weeks later, by which time the connection is invisible.

Standards relax. Busy rooms hide a great deal. The slightly-off things do not get corrected because nothing visibly breaks.

Costs settle at the higher level. ⚠️ Extra staff, extra stock, extra hours added for the peak quietly become the new baseline — and nobody reduces them, because reducing feels like retreating.

Problems get deferred. The awkward conversation, the supplier issue, the rota that is not working. All still there in the quiet month, now larger.

What to do during the good month

Keep feeding the pipeline at the same rate. ⚠️ This is the single highest-return discipline in the list and the hardest to maintain, because it is work that feels pointless while the room is full.

Bank the difference explicitly. Decide what the extra is for before it is absorbed. Unallocated surplus gets spent on the higher cost base by default.

Handle the deferred things now. You have slack you will not have in six weeks.

Write down what worked. ⚠️ In three months nobody will remember whether the strong month was the promotion, the weather, or an event down the road. Ten minutes at the time is worth more than any amount of reconstruction later.

Reading the quiet month correctly

Compare year on year, not month on month. Different question, different answer.

Check the pipeline six to eight weeks back. ⚠️ If it stopped during the busy period, you have found your cause — and it is a repeatable one.

Check whether costs came back down. Frequently the revenue is fine and the margin is not, which is a different problem with a different fix.

The pattern to break

Good month → relax → quiet month → panic → discount → thin margin → good volume → relax.

⚠️ The discounting step is the expensive one, and it is a response to a decline that was mostly variance plus a pipeline gap. Neither is fixed by cutting price.

Three to hold

Year-on-year by month. The only comparison that answers the question.

Pipeline activity by week, through busy periods. ⚠️ Shows the gap while it can still be closed.

Cost base before and after a peak. Whether the peak's costs went away when the peak did.

Where the record has to sit

Year-on-year comparison requires last year to be readable in the same shape as this year. Venues that change how they record things — or whose records live in a spreadsheet somebody rebuilt in March — cannot make the comparison that would have told them nothing was wrong.

tasteck keeps sales in one continuous record, so this month against the same month last year is a report rather than an excavation.

The quiet month is usually two things at once: ordinary variance, and a pipeline that stopped six weeks ago. Only one of them is worth acting on.