Fruit shop stocktaker - booking rules SOP for spoilage, markdowns, own use, and cut-fruit conversion - four exits, four ledger categories

You are a bookkeeping-rules consultant for a fruit shop. Write "the four
ways fruit leaves the shelf" as a booking-rules SOP, one page, followed
by manager, cashier and stock clerks alike. The premise: when the month's
margin won't reconcile, it is usually not missing fruit - it is four
different exits booked into one bucket. Spoilage, markdowns, own use and
conversion all heaped into "shrink" means no problem anywhere shows its
face.

The four exits (classify first, book second):
1. Spoilage: physically unsellable (mold, rot, off odor) - booked to
   "spoilage" at cost, photo attached. This category holds genuinely
   dead stock and nothing else
2. Markdown: still sellable, sold at a reduced price - the gap between
   original and actual price booked to "markdown loss," summarized by
   item. Note: the discounted revenue books as a normal sale; only the
   gap enters this category
3. Own consumption: eaten by the household, gifted, staff meals -
   booked to "own use" at cost, even a single melon. This is the
   owner-operator's murkiest account: unbooked, margin reads high and
   the month ends "profitable but broke"
4. Cut-fruit conversion: whole fruit becomes cut fruit - the raw
   material moves from whole-fruit stock to cut-fruit stock at cost;
   the cut sale books as cut-fruit revenue. Both sides must move:
   book the sale without the issue, and whole-fruit counts fill with
   variance forever; book the issue without the sale, and the cutting
   board works for free

SOP body:
- The four-exit spot test (one line each, in ink):
  physical state says spoilage (mold, rot, weeping, off smell)
  the price says markdown (sold, but cheaper)
  the pocket says own use (eaten here, or walked out as a gift)
  the form factor says conversion (it became cut fruit)
- The slip per exit: who writes it, when (on the spot / before close /
  weekly roll-up), and the format (item / quantity / amount / handled
  by) - one format everywhere, or the month-end roll-up dies
- Month-end summary template: four categories x amount x share of sales
  - the sheet handed to the owner; margin reconciles against it
- Sampling policy (its own line): whole fruit cut open for samples
  books to "own use" or to "marketing" - pick one, then never change
  it. Booked to neither, sampling is a bottomless hole

Prohibitions:
- No "record faithfully" / "clear the books daily" slogans - only
  classifications, slips and deadlines
- No fifth category - the finer the categories, the fewer get filled in;
  four is the ceiling
- No proposing new POS modules - written to run on paper slips or a
  phone notes template
- Once the sampling category is chosen it never flips - alternating
  months make both categories' numbers meaningless

Why it is written this way

Four separate categories solve one attribution problem: each exit points at a different operating lever - high spoilage points at storage and ordering, high markdowns at pricing and the freshness walk, heavy own-use at the family's generosity boundary, missing conversions at process. Heap them into one "shrink" bucket and month-end produces a sigh ("shrink is a bit high"); split them and every number names an action. The spot test uses the four crudest dimensions possible - physical state, price, pocket, form - so a clerk classifies in three seconds at the shelf; a test that takes a minute per item outlives no one's patience.

The markdown booking (gap into the category, revenue into sales) is the line newcomers get wrong first: booking the whole discounted sale as "markdown loss" understates sales; booking it all as sales hides the loss. The gap method keeps both ledgers pure - sales reflects real traffic and pricing, markdown loss reflects the price of clearing stock - and month-end both faces read true.

The own-use category exists for the owner-operator specifically: household and business drawing from one pocket is the accounting root of the "profitable but broke" illusion. Logging a single melon looks petty and is precisely what turns the category from "rough estimates" into "an actual ledger" - three months in, the owner sees for the first time how much fruit the household takes from the shop in a year, a number that matters to both pricing and cost.

"Sampling: pick one and never flip" is the smallest rule and the most quietly ignored: sampling is a daily trickle, and which category it lands in matters less than that it lands in the same one - flip it back and forth and both categories blur past reading. Half a rulebook's value is being right; the other half is being stable.

Variables