Lead-time demand is the stock that sells while you wait
Your daily sales multiplied by the lead time - from placing the order to the goods being sellable - is what will leave the shelf before the new stock arrives. At 40 units a day and a 14-day lead time, that is 560 units, and the page prints the figure as its own line between the two inputs and the reorder point so the multiplication can be checked without repeating it.
Daily demand sets the scale of everything after it. Enter units, not dollars: the reorder point is a shelf count, so it has to be expressed in the same unit your stock count uses. A figure drawn from the last few weeks tracks a real selling rate far better than a yearly average that flattens a seasonal peak into a number nobody actually sells at.
Lead time is just as easy to understate. It runs from the moment the order is placed to the moment the goods can be sold, so manufacture or packing, transit, customs and receiving all belong inside it. A lead time that forgets customs produces a reorder point that is too low in every single cycle.
Safety stock is the buffer on top
Lead-time demand assumes two things hold at once: that your forecast is right and that the supplier is on time. Safety stock covers the days when neither does - a late shipment, a demand spike, a picking error, a carrier that loses a carton. It is the only part of the reorder point that exists to absorb a surprise rather than to serve ordinary selling.
The buffer is an input, never a recommendation, and the page is explicit about that. Raise it after a stockout has actually cost you sales, and trim it when the holding cost of the surplus outweighs the risk it protects against. Two sellers with the same demand and the same supplier can reasonably choose different buffers and both be right.
More buffer is safer and costs more to carry, which is why the trade belongs to you. The page charges the buffer nothing and prices it nowhere; it adds whatever you enter to the lead-time demand and lets the total move with it.
The reorder point is the sum
Add the two figures and the result is the shelf level at which the next order should go out: 560 units of lead-time demand plus 120 units of buffer restock at 680. Crossing down through that line is the trigger - the page does not wait for zero, because the new order has not arrived yet and would leave a full lead time with nothing to sell.
The table prints the chain in order - daily units, lead time, lead-time demand, safety stock and the reorder point - so changing an input moves the whole route instead of producing an answer disconnected from the numbers that built it. The downloadable breakdown carries the same five lines for a spreadsheet.
What the figure does not answer is quantity. It says when to order, not how much you order: the order size is a separate decision about ordering cost against holding cost, and this page deliberately stays on the when side of it.
The rule flexes with how you review
The number is a trigger, and your rhythm of stock checks decides how much lead you build in. A shop that watches stock continuously can act on the reorder point directly. A shop that counts once a week should review against the reorder point plus what it expects to sell before that next count, because the shelf keeps selling through the gap.
There is no universal reorder point behind the page, only arithmetic on the three figures you supply. A 45-day ocean lead time at 25 units a day triggers at 1,275 units with a 150-unit buffer, while a local supplier on a four-day lead time at 60 units a day triggers at 320 - the same formula, two very different shelves.
The defaults are a worked case rather than a benchmark: 40 units a day, a 14-day lead time and a 120-unit buffer land on 680. A daily demand of zero stops the page instead of returning a trigger, because there is no selling rate to plan against, and a reorder point calculated for one product should never be carried over to another.
The average is not the risk: safety stock from variance
The buffer is the one number on this page with no formula behind it, and the honest way to size it is from variability, not from a rule of thumb. If the item sells 40 a day on average but the daily figure has a standard deviation of 6, a 14-day lead time needs about 1.65 × 6 × √14 ≈ 37 units of safety stock to hold a 95% service level - a quarter of the 120 the page defaults to.
Order size changes the picture too. If the supplier ships 400 at a time, average cycle stock is 200 units on top of the buffer, so on-hand at the moment you order is nearer 880 than 680. That extra stock is real money: at $4 a unit it is $800 tied up in the shelf, which is what the holding-cost page prices.
Reviewing stock weekly rather than continuously adds the demand you will sell between reviews - seven days at 40 a day is another 280 units. Reorder at 680 plus 280 and the order lands while the buffer is still intact; reorder at 680 alone and the review cycle eats the buffer before anyone notices.
The last constraint is the supplier minimum. If the minimum order is 1,000 units and the lead time sells 560, a single order covers about 25 days of demand - less than a month, which usually means the reorder point is the least of your problems: order quantity has become the inventory decision.
Danfio