It just can't tell you. These eight n8n workflows can — they compute reorder points from real consumption, find the capital frozen in dead stock, grade your suppliers on what they actually delivered, and message you before the shelf is empty.
One-time payment · yours forever · no subscription
The problem
Not because the data isn't there. Because a spreadsheet sits still and waits to be asked.
01 · Too late
The item hit zero days ago. Nobody looked. Now you're paying express freight, or losing the order — and the sheet knew the whole time.
02 · Invisible
Stock that hasn't moved in months doesn't announce itself. It just sits in the total, looking like inventory instead of money you can't spend.
03 · Unmeasured
They promised 12 days. They've been taking 16 for a year. You're still planning with 12, so you stock out on schedule.
Proof
This is the actual output from the included sample data — 120 SKUs and 6,695 stock movements. Not a mockup.
What's inside
Import the ones you need. They share one Google Sheet, so they compose — but nothing forces you to run all eight.
Measures 90 days of real consumption, computes safety stock and reorder point per SKU, and messages you the moment something crosses the line.
Subtracts reserved quantity, compares master stock against each sales channel, and pushes only what actually drifted. Ships in dry-run mode.
Classifies every SKU by value contribution and demand predictability, then totals the capital sitting in stock that hasn't moved.
Turns below-reorder items into supplier-grouped draft POs, respecting minimum order quantities and pack sizes. Drafts land in your inbox, not the supplier's.
Watches competitor pages for your key SKUs. Reads JSON-LD product schema first, so it survives site redesigns. Alerts only past your threshold.
Grades each supplier on real delivered lead time versus the promise, then tells you the lead time to actually plan with — average plus one standard deviation.
Daily snapshot of stock value in aging buckets, appended one row per day — so after a month it's a trend, not a number you argue about.
Ask your stock questions in plain language on Telegram. Sends only the most relevant rows to the model, so it stays fast and accurate on a 5,000-SKU catalogue.
How it works
No database to administer. No server to keep alive. Your data stays in a sheet you own, and every workflow is plain JSON you can read and edit.
Why this isn't another template dump
Across 13 warehouses, on spreadsheets, before writing a line of this. It shows in the parts that are easy to get wrong.
Reorder points use the standard continuous-review model, with a service level you can tune. ABC is genuine Pareto on turnover value; XYZ is the coefficient of variation of monthly demand. These are the formulas planners actually use.
Purchase orders are drafted to your inbox, never sent to suppliers automatically. Stock sync ships in dry-run mode and logs what it would have changed. You turn each one live only when you trust it. A workflow that emails suppliers on day one is how you end up with a duplicate order.
120 SKUs, 6,695 movements, 6 suppliers and a year of purchase orders — with dead stock, erratic demand and one deliberately unreliable supplier already in it. Every workflow produces meaningful output on the first run, before you touch your own data.
Plain n8n JSON and CSV files. No account to create, no API that can be switched off, no licence server to phone home. Edit the code, rewire the nodes, rename anything. One-time purchase — it keeps working whether or not anyone is still selling it.
What you get
Downloadable immediately after payment clears.
Before you buy
Self-hosted or n8n Cloud. Built and validated against n8n 2.x.
Google Sheets is the data layer. Free tier is fine.
A Telegram bot for alerts (free, 2 minutes). An LLM API key only if you want workflow 08.
Questions
You need to be able to import a workflow and select a credential. If you've built one n8n workflow of any kind, you're fine. INSTALL.md walks through the Google and Telegram setup step by step, including the two things people usually miss.
Workflow 02 uses a plain HTTP Request node rather than a platform-specific one, so any store with a REST API works. INSTALL.md has a worked example for Shopify and for WooCommerce — URL, auth header, body fields. Anything else needs those same three things changed.
The other seven workflows don't touch your store at all.
The workflows themselves scale fine. Google Sheets is the limit: it starts to drag past roughly 10,000 rows in the movements tab. INSTALL.md explains how to archive old movements — nothing in the pack reads them once they're outside your demand window.
Because the usual gateways aren't available to me, and I'd rather say that plainly than pretend otherwise. Payment is USDT on TRC-20. You send it, submit the transaction hash, and I email the download link — usually within a few hours, always within 24.
If that doesn't work for you, email me and we'll find another way.
Email me and I'll fix it or refund you. That's not a formal policy backed by a payment processor — it's one person answering their own inbox, which in practice means you get a faster answer and a real fix rather than a support ticket.
The sample data exists so you can see every workflow running before you invest time in your own data.
Yes. Installing and configuring the pack for clients as part of paid work is explicitly allowed. What isn't allowed is reselling or republishing the files themselves — on a template marketplace, inside another pack, or as a free download.
Selling your work is fine. Selling these files is not.
No. Workflow 08 is the only one that needs an LLM API key, and it's entirely optional. The other seven are ordinary automation with no AI and no per-run cost beyond what n8n and Google already cost you.
Neither. One payment, the files are yours, and they keep working offline forever. Fixes and improvements to version 1.x go out free to everyone who bought it.
One-time · yours forever · no subscription