Supply chain management: the moving parts a small operation actually tracks

Supply chain management is the work of getting the right things from suppliers into your hands and out to customers, at a cost and a speed you can live with. The term arrives loaded with enterprise software, and that framing does small operations a disservice: below a few hundred order lines a week, supply chain management is not a system you buy, it is five lists kept honestly enough that decisions can be made from them.

The lists are suppliers and what each one is actually good for, lead times as measured rather than as promised, orders placed and not yet received, stock on hand, and a record of what went wrong last time. Companies that run smoothly at this scale have all five and update them. Companies that do not have three of them, spread over two mailboxes and one person's memory.

What supply chain management covers, in order of how often it bites

• Lead time, measured. Not the number on the supplier website — the number you get by subtracting the date you ordered from the date it landed, averaged over the last ten orders, with the worst one written next to it.
• Reorder points. The stock level at which you place the next order, calculated from that measured lead time and your usage, so the decision is arithmetic rather than a feeling.
• Supplier concentration. Which items have exactly one source, and what the plan is for the week that source goes quiet. This is the single most useful column in the whole exercise.
• Open orders. What has been ordered and not received, with a date. Missing this list is why people order twice.
• Landed cost. Unit price plus freight, duty and handling. Buying on unit price alone is how a cheaper supplier turns out to cost more.
• Quality and incident history. What arrived wrong, from whom, how often. Without it you will renegotiate with a bad supplier on price.

Reorder points beat forecasting

Small operations are repeatedly sold demand forecasting and repeatedly disappointed by it. Forecasting needs history and stable patterns; when volumes are small, the noise is larger than the signal, and a sophisticated forecast is a confident wrong number. Reorder points are the practical alternative: hold enough to cover the usage you expect during the lead time, plus a buffer sized by how badly that lead time varies.

1. Take average weekly usage for an item over the last quarter.
2. Multiply by the measured lead time in weeks. That is your cycle stock.
3. Add a buffer for variability — a common starting point is half of the difference between the worst lead time and the average.
4. Set the reorder point at that total, and check it once a quarter rather than continuously.
5. Write next to it what happens if the supplier misses: substitute, second source, or tell customers early.

Track the worst lead time as well as the average, in its own column. The average tells you what to expect; the worst tells you what your buffer is really protecting you against. Teams that keep only the average discover the difference at the least convenient moment, and usually blame the supplier for a number they never recorded.

Where the lists live

Ettex Sheets is where the arithmetic belongs — usage, lead times, reorder points, landed cost — because these are calculations you want to see and adjust rather than a black box producing a recommendation. Ettex Records suits the supplier register and the incident history, since those are structured entries you filter and revisit rather than formulas. Purchase orders and supplier bills sit with Ettex Invoices, and the inventory side is covered separately in inventory management.

What we do not do belongs in the same paragraph. Ettex is not a supply chain platform: there is no EDI, no carrier or freight integration, no barcode scanning or warehouse execution, no demand forecasting engine, and no automatic purchase order generation. If you are running multi-echelon distribution or need customs documentation produced for you, this is the wrong shape of tool and a dedicated system will pay for itself.

The failure that is never about software

Almost every supply chain problem at this scale reduces to one of three things: a lead time nobody measured, a single source nobody flagged, or an order somebody placed twice because the open-order list was in an inbox. None of these are solved by buying something. They are solved by a person owning each of the five lists and a fifteen-minute weekly look at all of them, which is cheap and unglamorous and works.