Everything on this page is a barbell being made. The order from Iron Temple, the routing it goes down, the steel leaving the store, the hours at the CNC, and the number at the end: EGP 5,562.75 a bar. We have already loaded your products, your work centres and your people, so you can sign in and check it.
Your factory
and your books,
in one system.
Ten steps, seven desks, five apps — and one record. Heba takes the order, Karim raises the run, Dina buys the shortfall, Nour issues the steel, Fathy hands it down the line, Rania releases what passed, and Amr reads what a bar cost without asking anyone for a spreadsheet.


Iron Temple want eighty bars and three hundred kettlebells. I take the order here, and I can already see what is on the shelf and what has to be made — not because I rang the factory, but because the shelf and the factory are in the same system.
One order, two runs,
nothing retyped.
Iron Temple order eighty bars and three hundred kettlebells. Part of it is on the shelf; the rest becomes two production orders, each taking a copy of its routing. The copy is the point — Karim can change a step or a quantity for this run without rewriting the routing every other run uses.
What the run needs and what the store holds are the same two numbers, so the purchase order to Delta Alloy Steel is the difference between them, raised against the production order that asked for it. The steel is received at what it cost, and that is the layer the floor draws from.
Then the floor files documents rather than reporting progress. Material leaves the store against a step. A handover moves eighty shafts from the CNC to heat treatment and carries seventy-two hours of labour and overhead with them. A done production releases what passed inspection — and the order's produced quantity is that release, derived, never typed.
Nobody posts a journal. The nine shop-floor documents wrote EGP 890,040 of entries on their way past, and the invoice to Iron Temple knows the eighty bars cost EGP 445,020 to make — EGP 333,620 of steel and EGP 111,400 of hours.
A bill of materials
is not a parts list.
A bar is not steel plus bushings plus a carton. It is a bar that becomes a machined shaft, which becomes a heat-treated shaft, which becomes a chrome-plated shaft, which meets a sleeve assembly at the bench. Three of those exist only between two work centres, and every one of them has a cost.
So a routing here is a chain, and the app checks it is closed: an intermediate output no later step consumes is flagged before the order is released, because a broken chain does not fail loudly — it quietly never completes.
Scrap is added, not divided. Three percent saw kerf on the first step means the server stores 82.4 bars for eighty shafts — quantity × output × (1 + scrap) — and it stores it rather than recomputing it, so the figure on the document is the figure in the ledger.
It keeps your books. It has nowhere to put your factory.
This is not a complaint about QuickBooks. It records what you tell it, accurately. The problem is that half the cost of a barbell was never in it to begin with — it was in a spreadsheet beside it, maintained by someone, and reconciled once a year. Here is what moves across, and what changes once it has.
Figures you read, not assemble.
Every number below is from the demo company — fourteen production orders, nine routings, thirty steps, sixty-odd shop-floor documents. Nothing on this page was typed into a slide.
Produced against planned, per order
An order's produced quantity is derived from the releases against its final step. It cannot be typed, overridden or rounded up, which is why two of the fourteen orders read zero while material and hours sit at the bench.
Where the work in process is standing
Every withdrawal and handover names a work centre, so stock on the floor has a location and a step rather than being a single number called WIP.
What a unit cost, and what it was made of
Material and conversion stay separate all the way to the finished-goods value. A bar that cost more this month tells you whether it was the steel or the hours.
Work-centre absorption
Labor and overhead per hour per centre, against the hours shop-floor documents actually booked. A centre absorbing far under its budgeted hours is visible as a pattern, not as a surprise at year end.
Normal loss against abnormal loss
Saw kerf, swarf, casting risers and powder overspray stay in the cost of the good units. Porosity, weld cracks and out-of-tolerance weights leave work in process as a loss. Five reasons of one kind, six of the other.
The same figures as reports
Everything on the dashboards is also a report you can filter, export and schedule, so the numbers in a board pack are the numbers in the system rather than a spreadsheet built from them.
Built already, and being built for you.
- Three catalogues with their own identities — raw materials, work in process and finished goods — because a raw #5 and a finished #5 are not the same thing
- Multi-step routings where each step's output feeds a later step, validated before an order is released so nothing is stranded mid-chain
- Production orders that take a copy of the routing and stay editable, step by step and line by line, even when they came from a bill of materials
- The shop floor as documents: inventory withdrawals, center-to-center handovers and done productions, each moving stock and posting to the ledger as it saves
- Work centres carrying their own labor and overhead rates, with an absorption report over the hours actually booked
- Sales, purchasing, inventory and accounting against the same stock ledger and the same chart of accounts — no interface, no nightly export
- Dashboards and reports over all of it, filterable and exportable, in English and Arabic
- 01Migration from QuickBooks
Chart of accounts, customers, vendors, opening balances and stock on hand brought across as of a cutover date, with your first month run in parallel so you can tie the two out before you let go of the old one.
- 02Planned cost against actual, per order
The routing already computes what a run should cost. This puts what it did cost beside it, per order and per step, so an overrun is attributable to a centre rather than to the month.
- 03Capacity against the calendar
Budgeted hours per work centre against the hours already committed by released orders, so a delivery date you give Iron Temple is one the floor can actually keep.
- 04Per-seat permissions on the screens
The nine manufacturing permissions already exist on the server — stores staff raising a withdrawal without holding a manufacturing seat is the case that needs them. The screens gate per page today; this gates per feature.
Everything on the rest of this page is live in the product now. These four are the work we are quoting, and six to eight weeks is our estimate from the day you say go — the migration first, because nothing else matters until your opening balances tie.
Open the products and look around.
These are the five apps your flow runs across — the live products, not a sandbox copy. The demo company is Garage Gear: your barbells, kettlebells, bumper plates, hex dumbbells, rigs and plyo boxes, your ten work centres, your routings and your roster. Sign in with the account below and click through anything you like.

Routings, production orders, work centres and the shop-floor documents — withdrawals, center-to-center handovers and done productions.

Warehouses, receipts and the stock ledger every shop-floor document moves against.

Quotations, sales orders and invoices to gyms, boxes, clubs and distributors.

Purchase orders for steel, iron, rubber and coatings — raised against the order that needs them.

The ledger the floor posts into: work in process, conversion absorbed, finished goods, cost of sales.

Abdel Hamid will walk Garage Gear through this — and answer anything on the page that needs answering before you commit.
If the migration is what decides it, we can start on your chart of accounts and opening stock as soon as you confirm, and run your first month in parallel.