One Assembly, Three Stock Counts.

A small manufacturer we know sells a kit. Four parts, one box. The storefront says there are twelve kits in stock. The accounting system says there are twenty. The person on the floor says there are eight boxed and enough parts for about thirty more, if nobody sells the parts separately first. All three are honest. They disagree because nobody decided when a kit becomes a kit. Is it a kit when the parts are on the shelf, or when someone puts them in a box, or when a customer orders one?

That decision has a place in Zoho Inventory, and once it is made the three counts become one. Everything on screen is a mock demo of Zoho Inventory and of Light Manufacturing OS with sample data. Six minutes, one decision. In Zoho Inventory a product made from other products is a composite item. It has a bill of materials: the list of parts and how many of each go into one unit. Four parts for one kit. The parts are items with their own stock. The kit is an item with its own stock.

The question is how stock moves between them, and there are exactly two answers. You can bundle the kit at the moment it is sold, so the parts are consumed when the order ships. Or you can build the kit ahead of time, so the parts are consumed the day someone assembles them and the kit sits on the shelf as a finished good. Both are right for some businesses. What is wrong is not choosing, because then the storefront assumes one and the ledger assumes the other, and the floor does what it does.

Take the first choice and run the numbers. Say the shelf has forty of each part and no boxed kits. A storefront order for one kit arrives. Nothing happens to the parts until the order is packed. At packing, the system consumes four parts, one of each, and the kit ships. Before the order the kit count is zero and the parts are forty. After it, the parts are thirty-nine and the kit count is still zero, because the kit never existed as stock. So what does the storefront show as available?

The number of kits you could make: forty, the smallest part count. That is the honest number under this choice. The catch is what happens when a part is also sold on its own. Sell twelve of part two separately and the storefront's kit number drops to twenty-eight the same minute, whether or not anyone noticed. Now the second choice, same shelf. The floor builds ten kits on Monday. That is an assembly in Zoho Inventory: ten units of the kit come into stock and forty parts leave, four per kit.

Now the kit has a count of its own. Ten. The parts are thirty each. A storefront order for one kit takes the kit count to nine and does not touch the parts at all, because the parts were consumed on Monday. Under this choice the storefront shows the boxed kits, nine, not what could be made. That is the honest number here. The catch is the opposite one. If Monday's build is not recorded, the ledger still thinks there are forty parts and no kits, and the floor is holding ten boxes the system cannot see.

Which is exactly the manufacturer we started with. The second choice only works if recording the build is easier than not recording it. Here is how that looks in Light Manufacturing OS, the solution we design on Zoho for exactly this kind of company. This is the kit's page on the plant's own screens. The parts are listed with their stock. On the right is the pane, the AI control plane, and it knows this is the kit page. The planner types: build ten of these from the parts, and tell me what we are short.

The plane reads the stock of each part through Zoho Inventory's own connection, answers in the pane that there is enough for ten, and creates the assembly on her yes. The page refreshes: ten kits in stock, the parts down by forty. She could have pressed the page's own build button instead. Same result, same log entry. Talk or click on the same screen. Once the decision is made and the build is a one-line job, the last step is the connections. The storefront reads its available number from Zoho Inventory, not from its own count.

The accounting system, Zoho Books, reads the inventory value from the same record. And the floor sees the same kit page. Three counts become one because two of them stopped counting. That is the whole fix. It is not a bigger system. It is one decision about when a kit becomes a kit, recorded where it happens, and read by everyone else. If you want the same example as an explainer of what Zoho Inventory does end to end, that video is linked below. If your storefront, your ledger and your floor give three different numbers for the same product, the first question to answer is not which system is right.

It is when an assembly becomes a finished good. Answer that, record it where it happens, and the numbers agree. We work that out with the plant in a no-risk discovery: you pay only if you go ahead. The link is below.