Stock & Revenue Reconciliation

🔍 Stock & Revenue Reconciliation (v1 — open)

What this is

A from-the-logs reconstruction of how much stock we should have (every purchase in, every documented unit out) so that when you fill the Your physical count column, the gap = units that left without a clean record — off-log sales, personal use, friend trials, price variances. It also surfaces the revenue ledger gap so we can finalise how much the business actually earned and, from there, the founder ↔ company balance.

---

✅ Ground truth corrections (founder-confirmed, 2026-06-01) — do not re-ask

These override any conflicting record in the vault. The agent treats them as fact.

1. SKU-BPC157-5MG is NOT real — it’s seeded template/example data (it matches the Data Model sample record exactly: 42/6/20/18/89). The only BPC-157 we stock is 10mg. → remove the phantom SKU + cost card.
2. SKU-GHKCU-50MG is NOT real — same story. The only GHK-Cu we stock is 100mg. → remove the phantom SKU + cost card.
3. First BFF batch (PO-20251218-001, US$336) = ONE box / ONE vial. That’s it. It is not the source of any bulk stock. (Which single product/dose → confirm once; immaterial to totals.)
4. Wolverine Blend is self-made, not purchased. Founder combines 1× BPC-157 10mg + 1× TB500 5mg per blend. Landed cost RM79 = 40 (BPC10) + 39 (TB500) — the arithmetic confirms 1:1. Every Wolverine made/sold consumes one of each component vial from stock.

Part A — Stock reconciled to physical count (2026-06-01) ✅

Inputs: physical counts founder-supplied; W = 2 Wolverine blends made (1 sold, 1 given free) → −2 BPC-157 10mg & −2 TB500 5mg; first BFF batch = Retatrutide (BFF only ever sources Reta). R&D corrected 9→7 (see below).

Real SKU	Bought	R&D	→Wolverine	Should-have	Physical	Gap = sold/used off-log
Retatrutide 60mg	40*	—	—	26	4	22 🔴
GHK-Cu 100mg	10	—	—	8	0	8 🔴
BPC-157 10mg	10	—	2	8	5	3
TB500 5mg	10	—	2	8	6	2
CJC-1295 5mg	10	—	—	10	9	1
SS-31 10mg	10	7	—	3	3	0 ✅
MOTS-C 10mg	10	7	—	3	3	0 ✅
Wolverine Blend	(2 made)	—	—	0	0	1 sold + 1 free ✅

*Reta bought = 30 (PO-37066) + 10 (first BFF batch = 1 box of 10, founder-confirmed). 36 disposed of 40; gap 22 = disposed beyond the 14 documented.

What the gaps mean

• Reta 60mg — 22 vials gone off-log (on top of the 14 in the batch P&L; total disposed ≈ 36 of 40). This is the bulk of your unrecorded revenue.
• GHK-Cu 100mg — all 10 gone, only 2 ever recorded. Shelf is empty.
• BPC-157 10mg (3) · TB500 5mg (2) · CJC-1295 (1) — small off-log disposals.
• SS-31 & MOTS-C reconcile exactly once R&D is corrected: logs said 9 consumed, but 10 bought − 3 on hand = only 7 could have gone. Corrected STK-20260201-001/002 to qty 7. (If 2 were actually sold not R&D, tell me.)

Stock side is now real

Inventory on_hand updated to your physical counts; GHK-Cu 100mg flagged out, Reta/BPC/SS-31/MOTS-C low. The Bases now show reality. The variances above get booked as proper stock-moves + revenue once we price them (Part C).

---

Part B — Resolved / remaining

• ✅ W = 2 (1 sold, 1 free). ✅ First batch = 1 box of 10 Reta. ✅ R&D corrected 9→7. ✅ Reta price RM350→RM400.
• ✅ Money in = RM12,030 (personal account; total only, not a per-unit ledger).
• ✅ ORD-20260525-001 / TXN-20260525-001 (RM320, @ironpeak): confirmed fake seed data — deleted (see Part C3).

---

Part C — Rebuild true revenue from the units that left 🔴

The stock gaps tell us how many units left; you tell me the price each went for, and that is true revenue. The money ledger is effectively empty (only a suspected-fake RM320 entry), so we’re rebuilding it from physical reality. Here are the units to price:

Units gone (need price)	Qty	Price buckets to split into
Retatrutide 60mg	22	mostly @ RM400 (a few early @ RM350) · personal-use · free trial (RM0) · still-owed by Muz
GHK-Cu 100mg	10	@ RM90 / RM100 · personal · free
BPC-157 10mg (standalone)	3	price each · or free/personal
TB500 5mg (standalone)	2	price each · or free/personal
CJC-1295 5mg	1	price · or R&D/personal
Wolverine Blend	2	1 sold (RM350?) + 1 free (RM0)

Easiest way to answer

Don’t itemise every sale. Just give me counts per price bucket per SKU — e.g. “Reta: 12 @ 400, 4 @ 1000, 3 personal, 3 free.” I’ll total it into real revenue, real COGS (units × landed cost), and real profit, then book it.

Once priced, you also get the real margin per channel (dropship vs public) — which feeds the Pricing Model and the driver tree.

---

Part C2 — Sales ledger: who bought what (real, 2026-06-01)

Total money in to date = RM12,950, across four buyers. Muz = the entire bank statement (all credits are his, instalments now fully paid — no outstanding balance). The other three paid separately.

Buyer	Money in (RM)	Stock received	Paid vs free
Muz (main dropshipper)	12,030	~29–30× Reta 60mg + 1× Wolverine	all paid, settled
Kio (reseller, trial)	370	2× GHK-Cu · 2× BPC-157 10mg · 1× Reta · 1× MOTS-C · 1× TB500	paid only for 1 Reta + 1 GHK = RM370 (pre-pricing rate); rest free to test
Haikhal (reseller)	150	1× GHK-Cu	paid
Najiha (direct customer)	400	1× Reta 60mg	paid
TOTAL	12,950

Why we don't split Muz's RM12,030 by line

It’s a personal account and Muz paid in instalments, so the per-vial split isn’t bankable. What we know: Muz = RM12,030 for Reta (bulk, mostly RM400, a few early at RM350) + 1 Wolverine (RM350). That implies ~29–30 Reta vials.

Stock disposition — does it tie to the 40-Reta / full shelf? Mostly.

SKU	Disposed	Explained (paid + free + R&D)	Unaccounted
Reta 60mg	36	Muz ~30 + Najiha 1 + Kio 1	~4
GHK-Cu 100mg	10	Kio 2 + Haikhal 1	7 🔴
BPC-157 10mg	5	Wolverine 2 + Kio 2	1
TB500 5mg	4	Wolverine 2 + Kio 1	1
MOTS-C 10mg	7	R&D 6 + Kio 1	0 ✅
SS-31 10mg	7	R&D 7	0 ✅
CJC-1295 5mg	1	—	1
Wolverine	2	Muz 1 (sold) + 1 free	0 ✅

The big one: 7 GHK-Cu still unaccounted (only 3 of 10 went to named buyers). Plus small bits: ~4 Reta, 1 BPC, 1 TB500, 1 CJC. These left the shelf but aren’t in the RM12,950 — almost certainly personal use + free trials. Confirm and they get booked as owner-draw (personal) or marketing (free samples).

Real revenue — finally a number

• Revenue received to date = RM12,950 (Muz 12,030 + Kio 370 + Haikhal 150 + Najiha 400). No receivables outstanding.
• vs the RM320 the vault had booked — the books were under-recording by ~40×.
• COGS of paid units = RM4,387 → gross profit ≈ RM8,563.
• Separately, ~RM1,078 of stock at cost was given free or used personally (Kio’s free items RM180 + 1 free Wolverine RM79 + ~RM819 unaccounted/personal). That sits below gross profit — booked as marketing (free trials) or owner-draw (personal use) once you confirm the split.

---

Part C3 — Logs created & ledger tied out ✅ (2026-06-01)

Real records now exist in the vault and the full stock ledger nets exactly to your physical shelf for all 7 SKUs (verified):

SKU	Ledger (in−out)	Shelf
Reta 60mg · GHK-Cu 100mg · BPC-157 10mg · TB500 5mg · CJC-1295 · SS-31 · MOTS-C	=	=	✅ all tie

Created:

• 4 sales orders — ORD-20260303-001 (Muz), ORD-20260410-001 (Kio), ORD-20260415-001 (Najiha), ORD-20260515-001 (Haikhal).
• 12 revenue transactions — Muz’s 9 bank instalments + Kio + Najiha + Haikhal → money in = RM12,950 (reconciled).
• 13 stock-moves — first-batch intake (10 Reta) + per-customer outflows + Wolverine blend consumption + 5 adjustment moves for the untrailed/personal units.

Purged (deleted 2026-06-01) — all fabricated seed data, no longer in the vault:

Removed	Why it was fake
SKU-BPC157-5MG, SKU-GHKCU-50MG	products that don’t exist (matched template sample)
COST-BPC157-5MG, COST-GHKCU-50MG	cost cards for those non-existent SKUs
STK-20260120-000	phantom 10-Reta “intake from TCI” (TCI never supplies Reta)
ORD-20260525-001, TXN-20260525-001, FUL-20260525-001	the fake RM320 @ironpeak order + payment + shipment
TKT-20260526-001	support ticket for that fake order
LEAD-0001, CMP-001, POST-20260525-001, EXP-20260515-001	the made-up “Shred / TikTok campaign” narrative (lead, campaign, content, RM180 ad spend)
2026-05 month-close	fabricated numbers (rev 5,200 / net 2,500 / cash 7,400) — contradicted by the real reconciliation

Kept (real): 7 SKUs + 8 cost cards, the 3 real POs, real expenses (Lovable, Reta consumables), the Olymp competitor research, and the 4 customer orders / 12 transactions / stock-moves created today.

Analytics docs rebuilt from real figures (2026-06-01)

Mission & Policies, Value Architecture — Driver Tree (+ its interactive HTML) and KPI Dashboard have been rebuilt with the reconciled actuals: revenue RM12,950, gross profit RM8,563 (66%), inventory-at-cost RM1,437, resellers 97% of revenue (Muz 93%), TikTok/Threads RM0, cash commingled — pending founder↔company. The old fabricated figures (cash 7,400, rev 5,200, net-worth 12,289, ROAS 3.6) are gone.

---

Part D — Founder ↔ Company (after the above)

Cash-flow reconstruction (how much cash should be in hand)

Founder-confirmed model: the only money ever poured in was the first 2 batches (founder-funded). Selling that stock funded the 3rd batch (Reta 30) and everything since — the business has been self-funding from margin since the first batch sold.

Capital in (founder, one-time):

Source	RM
Batch 1 — Reta 10 vials (US$336 @ ~4.5)	1,512
Batch 2 — TCI group order (60 vials)	2,112
Total capital in	3,624

Cash flow since (with all real costs, founder-supplied 2026-06-01):

	RM
+ Revenue collected	12,950
− Batch 3 — Reta 30 peptides + consumables	4,285
− Janoshik COA testing (batch 3): 3.75 USDC + 50.14 USD	243
− Lovable website (Jan $25+RM61.10+Feb–May$5×4)	264
− Domain (RM124.04, 2-yr, renews 28 Jan 2028)	124
− Shipping/fulfilment (~12 × RM15 — estimate, confirm count)	180
= Cash generated (capital not yet returned)	≈ 7,855

The real number: ≈ RM7,855 cash generated + RM1,437 inventory = RM9,292 business value (on RM3,624 in → 2.56×)

Splitting it the way you framed it: your RM3,624 capital — already paid back to you ✅ · retained profit = RM4,230 (not RM4,935 — the missing ~RM705 is exactly the Janoshik/Lovable/domain/shipping costs that weren’t in the rough estimate).

Founder ↔ company — the actual answer

• Company owes founder: RM0 extra. Every cost above was paid from the same (personal) account that holds the revenue, and your RM3,624 capital is already returned. Nothing outstanding to you.
• Founder owes company: whatever you’ve drawn from the RM4,230 for emergencies. That retained profit is business money — meant to fund the next restock. To the extent it’s been spent personally, it’s a loan from the company to you.
• So “actual business money” = RM4,230 cash (physically there, or owed back by you) + RM1,437 inventory = RM5,667.

One number to close it: how much of the RM4,230 have you taken for personal emergencies? business cash still in the account = 4,230 − that. I’ll book the rest as an owner-draw (founder loan) so the books are clean.

Monthly cost of the business (now known) — strikingly low

Cost	RM/month	Type
Lovable (website)	~22.50	fixed (current $5/mo plan)
Domain (amortised)	~5.17	fixed
Fixed total	≈ RM28/mo
Shipping	RM15 × orders	variable
Janoshik COA	~RM243 per new batch tested	variable
Cursor / phone	not given	add if real

The business runs on ≈ RM28/month of fixed cost. That’s the real edge — it’s almost pure margin; the only meaningful costs are per-batch (stock + COA) and per-order (shipping).

---

What I need from you — checklist (v3)

• Physical counts · [x] W=2 · [x] first batch = 10-box Reta · [x] R&D fixed · [x] Reta RM350→400
• Money in = RM12,950 (Muz 12,030 settled + Kio 370 + Haikhal 150 + Najiha 400) · [x] no receivables
• The 7 GHK-Cu (and ~4 Reta, 1 BPC, 1 TB500, 1 CJC) not in any sale — personal use, or free samples? (Splits owner-draw vs marketing.)
• Muz exact Reta count (if you know it) — to nail the ~4 Reta personal gap.
• Company-paid-by-founder list: item · RM · date (for the founder ↔ company balance).
• Phantom records deleted (SKUs, cost cards, fake order/txn/ful, fake lead/campaign/content/ad, fake month-close) — vault now holds only real data.
• Analytics docs rebuilt with real figures (Mission, Driver Tree, KPI Dashboard).

Give me these and I produce v4: true revenue, real COGS & profit, the corrected net-worth figure, and the net founder ↔ company balance booked to a loan account.

---

Open items carried from existing notes

• Reta USDT overpay (~RM1,400): USD order line ($629 ≈ RM2,526) vs USDT leg paid (RM3,922.55) don’t tie — confirm what the extra covered before treating RM4,001 as final landed cost. (from PO-37066 / Batch P&L — Reta 60mg)
• GHK COGS booked against Reta batch: the 2 GHK-Cu sales sit inside the Reta batch P&L — book GHK’s own COGS separately for clean per-SKU margin.
• CJC-1295 DAC vs no-DAC: confirm which was shipped (price/retail depends on it). (from PO-20260120-001)

Related

• Capital — Playbook · SOP — Bookkeeping & Reconciliation · SOP — Monthly Close
• Mission & Policies (net-worth snapshot updates once this closes) · Value Architecture — Driver Tree