About

Why this exists

Replenishment tools are built for items with sales history. That is where the volume of decisions is, and they do it well. The first buy has no history by definition, so it falls back to a spreadsheet, two or three comparable products and a buffer. That method is reasonable, and it is missing the two things that decide whether it works: the spread around the estimate, and the fact that ordering too few and ordering too many cost very different amounts.

SKUZero formalises the comparison buyers already make. It takes comparable launches, yours, or typical patterns for your sector, keeps them as a distribution instead of an average, corrects the ones distorted by past stock-outs, and turns that into a first-buy quantity with the economics shown in the open.

Why it is free, and how it pays for itself

Because it costs almost nothing to run. There is no server, no database and no support burden, the whole thing is a static site that does its work in your browser, and the only running cost is a domain name.

There are no ads. Advertising at this kind of traffic would earn very little while costing exactly what this tool depends on: the confidence to put your own launch history into it. What is being evaluated instead is a paid tier for teams, a shared launch library, backtesting reports, branded exports, which you can look at on /for-teams. It does not exist yet; that page measures whether it should. The tool you are using now stays free either way, and nothing is being held back from it to create a reason to upgrade.

To be explicit about the obvious question: your data is not the product. Nothing you enter or upload is transmitted, pooled or sold. That is checkable in the network tab, and it is why the tool is built the way it is rather than as a normal SaaS with an account and an upload endpoint.

Who it is for, and who it is not for

It is built for buyers who commit to finished goods long before demand is known: technical and industrial distributors, importers and brand distributors of consumer durables, and private-label ecommerce sellers. What those three have in common is the decision, not the industry. One order, a supplier minimum, and a lead time long enough that the demand question is still open when the money is committed.

Four things it is deliberately not for, because pretending otherwise would waste your afternoon:

• Perishables. Shelf life turns leftover stock into a write-off on a clock, which is a different economics problem from the leftover recovery assumption used here.
• Fashion and seasonal apparel. Size and colour curves, markdown cadence and a hard end-of-season deadline dominate the answer. That is a well-served market with tools built specifically for it.
• Made-to-order and configured products. If you build against an order, there is no first buy to size.
• Enterprises running S&OP. If you already have a demand planning process, a statistical forecast and a planner who owns it, this adds nothing you do not have.

Four more things it does not model, all of them real, all raised by demand planners who reviewed it. Substitution: when this item is out of stock the customer often takes another one from you, so a stock-out costs less than the full margin. Cannibalisation: some of what the new line sells comes out of one you already stock, so it is less incremental than it looks. Project demand: orders that arrive in lumps because a specification landed, not because a curve says so. And range completeness: a line can be worth stocking to keep the range whole even where it does not pay for itself. Where those dominate the decision, the number here is an input to your judgment rather than a substitute for it.

Who made it

SKUZero is built and operated by Stepwyse bv, a Belgian company. It is a small independent project, built with buyers and demand planners at mid-size distributors in mind, HVAC, sanitary and building materials, technical distribution, electrical components. It is not affiliated with any ERP vendor, supplier or distributor, and it takes no commission on anything you order.

How it decides

Nothing is hidden behind a model you cannot inspect. Every recommendation shows the comparable launches it used, the assumptions applied and the arithmetic that produced the number, and the same inputs always produce the same answer. The method is written out in the guides.

Contact

Questions, corrections or a bug: info@skuzero.com.

Feedback from people who actually make these decisions is the most useful thing we can get, particularly where the tool is wrong, or where it asks for something you do not have.