Why reorders go wrong
A first program has everyone's attention: the sheet is written, the sample is measured, the confirmation is read. Reorders run on habit - and three habits break them. New staff re-quote from scratch instead of reopening the archived program, and small differences creep in. The approved sample disappears into someone's drawer, so there is nothing to check the new batch against. And the reorder point is set by feel, so the top sellers stock out exactly when the season peaks. All three are system failures with system fixes: an archive, a reference, and a trigger.
Archive these five things per SKU
A reorder is only as fast as the file it opens. Five records per SKU make any future reorder a one-message event, whoever sends it.
| Record | What it locks | Used when |
|---|---|---|
| Approved spec sheet, versioned | Every geometry, substrate and coating line | Every reorder, as the quoting basis |
| Golden sample reference | The physical geometry bulk is checked against | Receiving and dispute checks |
| Box and label artwork, versioned | Marking, barcode fields, origin line | Any repack or rebrand |
| Program confirmation | Quantity, lead time, papers, agreed in writing | Comparing this order to the agreement |
| Inspection results | What the last batch measured | Trending quality across batches |
If your program archive is currently an email thread, rebuilding it into these five records is an afternoon of work that pays back on the very next reorder. The sheet structure to archive is in the spec sheet template post.
Set the trigger against real lead time
A reorder point is a date calculation wearing an inventory costume. Work backward from when you need tools on the shelf: bulk production typically takes about 30 to 45 days after sample approval, a coated program adds a few days, and transit adds more depending on routing - every figure confirmed in writing per program on the MOQ and lead time page. Standard catalog items are the pleasant exception: boxed stock ships in days, so resellers filling from the catalog route can run leaner on those lines. The practical split: hold deeper buffer on ground-to-order SKUs, where the calendar is measured in weeks, and shallower buffer on catalog SKUs, where it is measured in days. A reorder point hit should feel boring - the order goes out with the archived sheet attached and nothing new is negotiated.
Signals that it is time to reorder
Beyond the arithmetic, four signals earn their keep. Consumption rate per SKU over a rolling window, not a gut feel from last month. Seasonality in your channel - some tooling categories breathe with construction or education calendars. Planned promotions or range launches, which create demand spikes no trailing average will see. And customer feedback on specific geometries, which is early demand data: a repeated question about a size you do not stock is a reorder in disguise. Programs that only reorder when shelves look empty are always negotiating from weakness; programs that reorder from data never think about it.
Split the range by velocity
Treat the range as three tiers with different mechanics. Top sellers - the few SKUs that move constantly - run on standing programs with firm reorder points, boxed route where possible, deepest buffer. The middle tier reorders on the trigger arithmetic above, quantities sized to sell-through. The long tail - the many SKUs that move slowly but define a full catalog - runs as consolidated top-ups on a fixed rhythm, so freight is spread across many lines instead of paid per hero SKU. The wholesale route exists for exactly this shape: small lots and boxed quantities fine, neutral boxes, quick turns. Trying to run the long tail on custom-program mechanics is the most common self-inflicted cost in a growing range.
Regrind as part of the plan
Reorders are not the only way inventory comes back. Worn solid carbide tools can be re-ground on the original geometry and re-coated, quoted per batch and confirmed per program - a sensible complement for expensive geometries and for customers who run the same profiles constantly. Whether regrind earns its place in your program depends on your customers' economics, but deciding deliberately beats discovering the service exists when a customer asks. The sharpening and regrinding landscape is covered in the sharpeners and regrinding post.
Make the next reorder one message
The test of a reorder system is its message count. A healthy one reads: "Reorder per program sheet v1.3, SKU list attached, same marking and packing." The desk replies against the archive - same geometry, same golden sample reference, same papers - and the only open questions are quantity and timing. If your next reorder cannot be sent as one message, the archive is missing one of its five records. To open a program designed for reordering from the start, send the range and the sell-through you expect through the quote form, and the buyer support team will set up the program records with you.
Working the calendar backward, a sketch
An example makes the arithmetic concrete - illustrative figures, not promises. Say a top-selling SKU needs tools on the shelf by a season start. Working backward: transit on top of production, bulk at about 30 to 45 days after sample approval, sample approval after a sample round of about 15 to 25 days for custom ground tools, and the program confirmation before that. Add a buffer for coating days and the ordinary friction of calendars, and the order belongs in the desk's queue a season ahead - which is precisely what the archive plus reorder point is designed to make routine. Standard catalog SKUs run the same sketch against a much shorter clock, since boxed stock ships in days; that difference in clock speed is the deepest reason to split the range by velocity rather than managing it as one inventory.
Keep your own lead-time history
One more calendar habit: record actual dates on each completed program - ordered, sampled, approved, shipped, received. Three reorders in, you hold your own empirical lead times per lane and per program type, and the reference ranges become a floor you plan on rather than a ceiling you hope for. That history is also the calmest possible evidence in any timing conversation, because it is your data about your programs, not anyone's promise about someone else's.
Practical checklist
- Archive five records per SKU: sheet, golden sample, artwork, confirmation, inspections.
- Version everything; name the version on every program confirmation.
- Set reorder points from real lead time: about 30-45 days bulk, plus coating and transit.
- Run top sellers on standing programs; run the long tail on consolidated top-ups.
- Watch consumption, seasonality, planned promotions and customer questions.
- Decide deliberately whether regrind and recoat belongs in your program.
- Aim for reorders that fit in one message, against the archive.