Why the inventory ledger does not tie, and how to find the gap

The trial balance says raw materials inventory is 184,212.40. The stock valuation report says 183,800.40. The gap is 412.00, small enough to tempt a plug entry and large enough that the auditor will ask what it is. The hunt is slow because the gap has no label: in most systems nothing tells you whether it is a journal, a late freight bill or a standard that changed without a trace.

Three usual suspects, none of them labeled

The first suspect is a manual journal posted straight to the inventory account. The general ledger accepts it, the stock report never hears about it, and the two drift apart by exactly that amount. The second is a landed-cost bill that arrived after the goods had already sold, so part of the cost it adds has no stock left to sit on. The third is a change of standard cost made on the item record without a matching revaluation, which leaves the old and new standards both implied in the history.

Legacy tools make all three look identical: a difference with no author. A controller ends up exporting both sides to a spreadsheet, matching by item, and reading document histories one at a time. The work is real accounting judgment, but most of the hours go to finding where to look.

Why the two sides agree to begin with

In Cairn the stock valuation and the inventory account are written from the same posted documents. A receipt, a shipment, a build or a work-order completion moves the cost layer and posts its general ledger legs together, so ordinary activity cannot open a gap between them. Anything that does open a gap has an author: a manual journal on the inventory account, landed-cost residue, or a standard changed on an item before revaluations were required.

That is also why parallel costing matters to the close and not only to margin analysis. FIFO, LIFO, weighted average and standard cost are kept at the same time for every item and lot, and each receipt is its own cost layer, so asking how an item would read under another method never means rebuilding its history.

A freight bill that lands after the sale

Take the second suspect. A pallet of 72 cases is received, 54 cases ship in the same month, and the ocean freight bill arrives the following week. The bill adds cost to a layer that is now mostly gone. The tidy-looking fix, spreading the added cost across everything the item has ever held, rewrites history and moves margin on orders that already invoiced.

Cairn's landed-cost handling does something plainer. The part of the true-up that belongs to the 18 cases still on the shelf goes into their layer. The part that belongs to the 54 cases already sold is expensed as it is recognized, in the current open period, and not smeared back across closed ones. The inventory account and the valuation report move by the same amount on the same day, and cost of sales shows a visible, dated line you can point to.

A change of standard is a document

The third suspect is the quietest. Someone updates a standard cost because a supplier raised prices. In many systems that is an edit to the item, and the value of the stock on hand silently changes meaning: the inventory account still holds the old standard's value while the report prices the same units at the new one.

In Cairn a change of costing method or standard is itself a dated document, an inventory cost revaluation numbered ICR-n. It has its own postings, it is dated in the current open period, and it is never an edit to the item. The revaluation drains the stock at the cost it carried and pushes it back at the new standard, with the difference landing in Cost of Sales Adjustment. The change has an author, a date and a number.

So two implied standards with no revaluation between them can only be history from before revaluations were required: an item whose standard was edited in place. Cairn refuses that edit today, and the diagnosis below still names the old case, so the controller can see where the history and the ledger part.

What Flint does with the question

Ask Flint, the assistant, why the inventory account and the valuation disagree and it runs Cairn's costing diagnosis, which reports the tie per account, as of a day. For each account it names the sources of any difference: a manual journal on the inventory account, landed-cost residue, two implied standards with no revaluation between them. It also says which items are drifting and from what, whether the costing method comparison is stale, where the replay floor sits and which months are closed, and when the latest cost-of-sales catch-up ran.

Flint reports what it found and where, built from the ledger and the reporting views. The controller decides what to do with each source: whether a journal belongs on the inventory account, whether residue should be cleared, which document explains an old standard.

Flint does not post the correction. A read needs no approval, and a fix is a record you review and approve like any other, under your role. The diagnosis is cheap to run on any day you like, so the question gets asked on day two of the close, when the gap is one document wide, and not on day four, when it has been joined by a dozen more. The accounting and finance view shows the same close from the controller's side.

What the close looks like

A named source moves the gap from the hunting stage to the deciding stage, where a controller's time is worth something. On the financial close board the inventory account is explained, not forced: 412.00, two sources, JE-2207 and the landed-cost residue on PO-2236.