Most seasonal brands plan Q4 forwards. They decide in September what they want November to look like, then find out in October what is actually possible. Doing it in the other direction, in June, changes what you can promise.
Count backwards from the cutoff
Start with the last day Amazon will deliver for the holiday. Then subtract, in order:
- Two to three weeks of FBA receiving time in peak, which is considerably longer than the same process in April
- Your freight transit, which for sea from Asia or southern Europe is four to seven weeks depending on the lane
- Factory lead time, which is also longer in autumn because everyone else is doing this too
- Two weeks of slack, because something will go wrong and this is the cheapest insurance available
For a cookware client manufacturing in Portugal, that arithmetic produced a purchase order deadline of 4 July. Not September. They hit it with two weeks to spare and did not go out of stock on a hero product for the first time in four years.
Decide what you are willing to sell out of
You will not fund everything. Rank the catalogue by contribution per unit, not revenue, and fund the top of that list to full cover. Deliberately under order the bottom and accept that it sells out. Going out of stock on a low margin product in mid December is a choice you can make on purpose in June, or a thing that happens to you in December.
Protect the rank you are about to buy
The expensive part of running out is not the lost sales. It is that the organic rank you spent all year building collapses in January and takes until April to recover, so you pay for it twice.
The cheap version of insurance is a small always on budget through the quiet months against the terms you need to own by October. One client spends about $1,200 a month through the spring for exactly this. It means they enter Q4 already ranking rather than buying their way back up in the most expensive eight weeks of the year.
Write the plan down
Purchase order dates, freight bookings, the budget by week from October to December, and what you will do if a container is late. The value is not the forecast, which will be wrong. It is that in November everybody already knows what happens next, and nobody is on the phone to a freight forwarder in the second week of December.
