SaleShade Operations · Stock audit

Physical stock against the books, without shutting down

Blind counts against the books, without shutting down — gains and losses shown separately, and the item that keeps varying surfaced to you by name.

The count that happens once a year, if it happens

Every owner knows the number in the books and the number on the racks drift apart.

What stops him from checking is the price of checking: a proper count means shutting down — so it happens once a year at closing, in a hurry, with everyone tired.

And when it does happen, the counting has a quiet flaw: the counter knows what the book says. A man counting toward a known number finds that number more often than he should.

That is not an audit; it is a confirmation exercise.

Scope the count, and the business keeps running

An audit here covers what you choose: everything — the true shutdown, usually at financial year closing — or a slice: a category, a brand, or items picked by hand. Scoping is what makes auditing routine instead of annual.

One honest cost, stated plainly: items under count are locked from allotment, and orders needing them wait.

The lock exists because the classic failure of every physical count is somebody picking from the rack mid-count — after which the variance you find is an artefact of your own process.

Count everything

The true shutdown, usually at financial year closing. The whole business stops, and you find out what the racks actually hold.

Or count a slice

A category, a brand, or items picked by hand. Small scopes keep the lock short, and the shop keeps running — which is why scoping matters.

Small scopes keep the lock short; that is exactly why scoping matters.

The counter never sees the book figure

Counting toward a known numberThe old way
  • The counter knows what the book says — and finds that number more often than he should
  • He finds it because he expected it — which is the whole flaw in a sighted count
Counting blind
  • The book figure is not on the screen — the warehouse team records what is physically there
  • Photographs are mandatory — the only place in the platform where an image is not optional
  • Because a mistaken item is invisible — recording A against B corrupts two lines while looking correct

In an audit, recording item A's count against item B corrupts two lines while looking perfectly correct. A quantity error is recoverable; a mistaken-identity error is invisible.

The photograph is what makes it visible.

Gains and losses are shown separately, on purpose

After review — typed against the books, or bulk-imported from the ERP — the audit produces its result: total gains in green, total losses in red, net at the end, in both count and value, with a running position across audits.

Audit result
Gains +22Losses −49Net −27
Soft-close hinge 110°+18
Drawer channel 450mm-18
Cabinet handle 192mm+4
Tower bolt 6"-31
The first two are a pairing — A was billed, B was dispatched. Collapsed into one number it would read as zero.
Illustrative figures. Gains and losses are never collapsed, because the pairing is the thing worth seeing.
A pairing points upstream

A was billed and B was dispatched, so one gains exactly what the other loses. That is picking and checking — not theft.

A loss with no offsetting gain

The different, darker signal. The system does not promise zero discrepancies; it tells you which kind you have.

The item that keeps varying gets named

One variance is noise. The same item varying across three audits is a signal — and it is a signal nobody catches by hand, because each audit is settled and forgotten.

Here, items with repeated variances across audits are surfaced to you separately, by name.

Every count leaves a stamp

Scoped, counted, photographed, reviewed, settled — who and when, to the minute.

Discrepancies settle into your books by voucher, as receipts do.

The act itself makes the stamp.

An audit you can afford to run in March, and again in July

Scoped counts, blind entries, photographs that prevent the invisible error, and results that tell you what kind of problem you have — while the shop stays open.

Scope one brand, run one blind count, and read what the racks actually say.

Ask for a demo