01The second system

Inventory thatforecasts oneitem at a time

A standalone, multi-industry app on its own AI-first data model: forecasting, reorder points, lot traceability and valuation, without the shop-floor surface.

Running a plant, not a warehouse? See the Manufacturing platform. Already sold? Jump to pricing.

03Not a mockup

The actual app, a real tenant, zero staging

app.inventoryanalytics.ai/dashboard
Inventory Analytics dashboard: 2000 SKUs tracked, 18 low-stock, and a live work queue of dead-stock and overstock findings
app.inventoryanalytics.ai/items
Items table with real SKUs, unit cost, unit price, on-hand quantity and reorder point
app.inventoryanalytics.ai/analytics
Analytics advisor feed narrating dead-stock and overstock findings per SKU

04Forecast, then reorder

A part that moves twice a quarter isn't fitted like one that moves daily

The model is chosen per item, from that item's own consumption history.

  • IntermittentCroston's method · SBA bias correction

    Long gaps, then a movement — fitted on the gaps as well as the sizes.

  • TrendHolt

    Demand drifting one way, with level and slope tracked apart.

  • SeasonalHolt-Winters

    A shape that repeats — the season is fitted, not smoothed away.

  • AverageAverage

    No trend and no season, so nothing is invented to fit one.

Then the reorder point

Safety stock and reorder points are derived from the forecast and from the variance in supplier lead times, then drafted as a purchase order per supplier.

reorder → draft

  1. demand_forecastper item, from its own consumption history
  2. lead_timeper supplier, and the variance in it
  3. safety_stockderived from both
  4. reorder_pointderived from both
  5. po_draftone per supplier, off the result
  6. confirmationa person — always

05What the ledger has to survive

Two things a stock system is judged on later

Lots, recalls and cold chain

Per-item shelf life, first-expiry-first-out consumption, storage-condition fields, and recall genealogy that traces a lot back through the supply chain.

  1. Supplier
  2. Lot received
  3. Storage
  4. FEFO
  5. Where it went

Valuation on a ledger models can read

FIFO, LIFO and weighted-average costing, aging and turnover — computed over a transaction ledger designed from the start to be read by models, not just rendered in a table.

  • FIFO
  • LIFO
  • Weighted average
  • Aging
  • Turnover

Built for

  • Apparel
  • Food
  • Manufacturing
  • Pharma
  • Hospital
What each one gets, in detail →

06Outside signals

Demand doesn't start in your warehouse

It reads the WHO's Disease Outbreak News feed and Google Trends, then matches what it finds against your own catalog.

  1. 01WHO Disease Outbreak News · Google TrendsTwo public feeds, read on your behalf.
  2. 02Matched against your catalogNot a headline — the item you actually stock.
  3. 03Surfaced as a recommendationIt arrives in the app as a suggestion, nothing more.
  4. 04A person confirms — or doesn'tThe system never places the order itself.

It never places the order — every signal arrives as a recommendation for a person to accept or ignore.

07Inside the product

What ships today, not on a roadmap

  • Per-item forecasting

    Croston + SBA for intermittent demand, Holt for trend, Holt-Winters for season, average otherwise — chosen per item.

  • Reorder points, gated

    Safety stock and a purchase-order draft per supplier. Nothing sends itself — a person confirms every write.

  • Lots, FEFO, recall

    Shelf life, storage conditions, and a recall genealogy traced backward through the supply chain.

  • Ledger-grade valuation

    FIFO, LIFO, and weighted-average costing, plus aging and turnover, on a ledger built to be read by models.

  • Outside demand signals

    WHO Disease Outbreak News and Google Trends, matched against your own catalog — a recommendation, never an order.

  • Five industries, one model

    Apparel, food, manufacturing, pharma, and hospital supply, all on the same forecasting and lot data model.

08Five industries

Same ledger, five different things it has to survive

  • Apparel

    Per-item forecasting for size and season curves, without a generic reorder rule flattening the shape.

  • Food

    Shelf life, FEFO consumption, storage-condition fields, and recall genealogy traced back through the supply chain.

  • Manufacturing

    Lot and valuation data that reads the same whether the unit is a raw material or a finished good.

  • Pharma

    The same lot and recall machinery food runs, held to the traceability a regulated shelf needs.

  • Hospital

    Supply tracking on the forecasting and reorder-point model, not a spreadsheet someone owns personally.

09Pricing

Start free. Add the model when tracking alone runs out.

One app, three tiers. Every button installs the same Microsoft Store listing.

  • Free

    Track items, lots, and locations. No card.

    $0

    Get it on Microsoft Store
    • Items, locations, and lot tracking
    • CSV import
    • Low-stock alerts
  • Pro

    Valuation and outside signals for the full ledger.

    $69.99/mo

    Get it on Microsoft Store
    • Everything in Growth
    • FIFO / LIFO / weighted-average valuation, aging, turnover
    • Outside demand signals — WHO outbreak feed, Google Trends

10Asked about this product specifically

What people check before they trust a forecast

  • 01Does it place orders on its own?

    No. Reorder logic drafts a purchase order per supplier, and an outside signal arrives as a recommendation — but drafts stay drafts until a person confirms them. The same gate covers every write the chat assistant proposes, from a stock adjustment to a purchase order.

  • 02Which forecasting methods does it actually use?

    Croston's method with the SBA bias correction for intermittent demand, Holt where there's a trend, Holt-Winters where there's a season, and a plain average where there is neither — chosen per item from that item's own consumption history rather than set once for the whole catalog.

  • 03Do we need the manufacturing platform to run this?

    No. Inventory Analytics is standalone, on its own AI-first data model. If you are running a plant rather than a warehouse, the manufacturing platform is the other door.

  • 04Is there a free tier?

    Yes. Free tracks items, lots, and locations with CSV import and low-stock alerts, no cost. Forecasting and reorder points are Growth; valuation and outside signals are Pro. All three install from the same Microsoft Store listing.

Anything else, ask it directly — the Manufacturing page covers how the modules share one data model.

11Get the app

Get Inventory Analytics

Install it on Windows and point it at one catalog or one season of consumption history — the forecasts and reorder points are ready as soon as the data is.

REQ · 001 · DEMO

Book a demo

Twenty minutes on your own data — a BOM, a stock ledger, or a week of downtime. Tell us where to send the invite.

We'll only use this to get in touch about a demo.

✓

Thanks — we'll be in touch.