Inventory Forecasting — Ask the Vendor to Show You the Formula
Cogsy is $199 a month. Sage's Inventory Planner won't publish a price. Both will tell you what to reorder; neither shows you the arithmetic. Here is why that matters, and the four formulas the whole category is built on.
Demand forecasting software makes a specific kind of promise: tell us your sales history and we will tell you what to order.
It is a good promise. Running out of a best-seller costs you the sale and sometimes the customer; over-ordering ties up cash in a stockroom. Somewhere in between is a number, and working it out by hand across four hundred SKUs is genuinely hard.
What almost nobody in this category will do is show you how the number was reached.
What it costs
Cogsy — from its own pricing page (read 2026-07-05):
"All in one — $199/month" (after a 14-day free trial)
Inventory Planner (by Sage) — from its own pricing page (read 2026-07-05):
No public price — quote-gated.
Cogsy at $199/month is $7,164 over three years. For a small retailer or e-commerce shop, that is real money — and it buys, in essence, an arithmetic service.
The arithmetic in question
Here is the uncomfortable thing about this category: the core calculation is not a secret and it is not complicated.
The industry-standard reorder point is:
reorder point = (average daily demand × lead time) + safety stock
That is it. That is the formula almost every inventory-planning tool in this price bracket is built around. The sophistication in expensive systems is real — seasonality, promotions, supplier variability, multi-warehouse allocation — but for a shop with steady demand and one supplier per line, the equation above is most of the answer.
Which raises the question the category prefers you did not ask: if the maths is standard, what exactly is the $199 buying? The honest answer is: the data plumbing, the interface, the alerts, and somebody else's responsibility for keeping it running. Those are worth money. They are just not the same thing as proprietary intelligence, and the pricing implies the latter.
The self-hosted option, and its one genuinely unusual claim
Restock is self-hosted inventory forecasting — reorder points and suggested order quantities per SKU. $89.00, one time (product page, read 2026-08-01).
Its answer to the black-box question is the reason to write this article:
"Is the forecasting math a black box?" — "No. Reorder point equals average daily demand times lead time plus safety stock — the four formulas are printed in the app and the README, and covered by a 140+ assertion test suite."
Printed in the app. Not "explainable AI," not a methodology whitepaper behind an email gate — the actual equations, visible next to the numbers they produce.
That matters for a practical reason, not an ideological one: when a forecast tells you to order 340 units of something and your instinct says 200, you need to know which input is driving it. With the formula visible you can see that lead time is set to 45 days when your supplier now ships in 20 — and fix the input. With a black box you can only override the output, every time, forever.
Two details that suggest it was built by someone who has done this
"How do I get my sales data in?" — "Export orders from your POS or cart as CSV with date, SKU, and quantity columns — names and order don't matter, Restock maps by header. Bad rows are skipped and reported line-by-line."
"Reported line-by-line" is the tell. Every real import of real retail data has bad rows. A tool that silently drops them produces a forecast quietly built on 94% of your sales.
"Does it handle new products with no sales history?" — "Yes — no divide-by-zero, days-of-stock shows as unlimited, and you can set a manual safety-stock floor in units until history accumulates."
New products with no history are exactly where naive forecasting tools produce nonsense or crash. Saying "no divide-by-zero" out loud is a small thing that suggests somebody hit it and fixed it properly.
What you are giving up
No live integration. You export a CSV from your POS or cart and import it. Cogsy and Inventory Planner connect directly to Shopify and similar — that is a genuine convenience and for a busy shop it may be decisive. A weekly manual export is a habit somebody has to keep.
No purchase-order sending. Forecasting tells you what to order; you place the order.
No multi-warehouse allocation, no promotional modelling. If you run campaigns that spike demand tenfold for a week, a standard reorder-point model will not anticipate them and neither will your instincts alone. You will need to override, and you should know that going in.
No supplier-side data. Lead times are what you enter. If your supplier's actual lead time drifts and you do not update the field, the forecast drifts with it and nothing warns you.
Who should use which
Cogsy — you run e-commerce at volume, want live integration with your store, and $199/month is small against your inventory value. The 14-day free trial is published; use it before deciding anything.
Inventory Planner (Sage) — you are already in the Sage ecosystem or large enough that a quote is normal. Ask for the number in the first email.
Restock — you have a few hundred SKUs, steady-ish demand, one supplier per line, and you are currently doing this in a spreadsheet or by memory. And you want to be able to see why a number is what it is.
A spreadsheet — under about fifty SKUs. The formula is one line; a spreadsheet computes it fine, and the tool earns its place at the point where maintaining four hundred rows by hand stops being realistic.
The honest costs
Garbage in, garbage out — and here it is expensive. A forecast built on an incomplete export or a stale lead time will confidently tell you to buy the wrong things. Check the first month's suggestions against your own judgement before you trust any of them.
The export habit. A weekly CSV is not hard; it is just a thing that must actually happen.
Backups. Less critical here than in a financial register — the source data lives in your POS — but the settings and lead times are yours.
Hosting. "PHP 8+ with PDO — present on virtually all cPanel/Plesk hosts." For a concrete number from a named provider: Hetzner's own pricing — machine-read from the price API behind their pricing pages on 10 Aug 2026 — lists web-hosting plans from €1.60 to €16.72 a month and entry cloud servers from €3.99 a month (prices as returned for their German datacenters; confirm current pricing and VAT treatment at checkout). Any comparable host works — the point is that the "own" side of this comparison costs single-digit euros a month.
The short version
Inventory forecasting is $199/month published, or a quote. The core arithmetic — average daily demand × lead time + safety stock — is standard and not secret.
So the question to ask any vendor in this category is: will you show me the formula? A tool whose maths is printed on the screen lets you fix the input when the output looks wrong. A black box only lets you disagree with it.
If you need live store integration and PO sending, pay for the platform. If you need the number and the reasoning, that is $89 once — and under fifty SKUs, a spreadsheet is still the right answer.
Next steps
- Open the live Restock demo — no sign-up, already signed in, resets itself
- Restock product page and licence options
- The SaaS Price Observatory — Cogsy, Inventory Planner and 80 others, quoted from their own pages and dated
Prices are quoted from each vendor's own pricing page with the date read. The reorder-point formula stated above is standard inventory-management practice, not a proprietary method of ours or anyone else's.