Getting started

Three columns is all you need. Everything else is optional and simply unlocks more of the app.

The minimum file

One row per sale, or one row per product per day. Both work — the app adds up whatever it finds.

ColumnWhat it isExamples of what it might be called
DateWhen the sale happened Date Invoice Date SalesDate trandate BEDAT
ProductThe product name or code Product Description Item SKU MATNR
QuantityHow many units sold Quantity Qty Units Sold MENGE

The app recognises hundreds of column names across English, German, Spanish, French and SAP field codes. If it guesses wrong, you can correct it from a dropdown — nothing needs renaming first.

A valid file looks like this

Date,Product,Quantity Sold
2024-01-03,Cotton T-Shirt White,47
2024-01-03,Instant Coffee 200g,62
2024-01-04,Cotton T-Shirt White,51

Optional files — each unlocks more

Add thisColumns neededWhat it turns on
Stock levelsProduct, quantity on hand Reorder alerts, overstock warnings, stock turnover
CostsProduct, purchase price Every figure in money rather than units, plus proper order quantities
Purchase ordersSupplier, order date, receiving date Real supplier lead times and reliability, instead of an assumed number

Drop them all in together. Each file is identified automatically — you do not need to say which is which.

Exporting from common systems

Tally (ERP 9 / Prime)

  1. Gateway of Tally → DisplayAccount BooksSales Register
  2. Press F12, set Format to Detailed so item lines are included
  3. Set the period to at least the last 12 months (Alt+F2)
  4. Alt+E to export, choose CSV or Excel

Tally writes dates as DD-MM-YYYY. The app detects this and shows you what it decided — check it on the Data check tab.

Excel or Google Sheets

  1. Make sure row 1 contains column headings
  2. Remove any subtotal, blank or "Grand Total" rows
  3. Delete merged cells — they break CSV export
  4. File → Save As / Download → CSV

If your file is over the Excel row limit of 1,048,576 it has been silently truncated. The app detects that and warns you.

Zoho Books / Inventory

  1. ReportsSales by Item
  2. Set the date range to 12–24 months
  3. Group by Item and Date rather than by month
  4. Export AsCSV

QuickBooks

  1. ReportsSalesSales by Product/Service Detail
  2. Set the report period, then Customise → include the Date column
  3. ExportExport to Excel, then save as CSV

SAP

  1. Transaction VA05 (orders) or MB51 (material movements)
  2. Include MATNR, MENGE and a date field
  3. Export via List → Export → Spreadsheet

SAP's field codes are recognised directly — no renaming required.

A POS or till system

  1. Look for Sales report, Product mix or Item sales
  2. Choose the longest period available, ideally two years
  3. Export detail rather than a summary — daily lines, not monthly totals

How much history to include

PeriodWhat you can rely on
2 years or moreSeasonality confirmed by repetition. Order quantities are dependable.
1 yearSeasonality seen once. Reasonable, not yet confirmed.
6–12 monthsRankings and cash figures are sound. Annual quantities are projected.
Under 6 monthsUseful for spotting priorities and stuck stock. Not a basis for ordering.

The app grades your file on exactly this and states plainly what it can and cannot support — before showing you any numbers.

Common problems

SymptomCause and fix
"No sales history found" The app could not identify three required columns. Open the mapping dropdowns in the Data panel and set them by hand.
Dates look wrong A file of 03/04/2024-style dates with no value above 12 is genuinely ambiguous. The app says which it assumed — override it in the Data panel if wrong.
Many rows skipped Usually subtotal rows, blank lines or text such as "N/A" in the quantity column. The Data check tab itemises exactly what was dropped and why.
Numbers look 1000× off A European file read as Anglo, or the reverse — 1.234 means 1,234 in one and 1.234 in the other. The Data check tab shows which was used.
"Model adds 0% over a simple average" Not a fault. Your demand has no learnable pattern, and the app is telling you so instead of pretending otherwise. Use the historical average and a stock rule.
Large file is slow Files over about 100 MB take a minute or two. Everything is computed on your own machine, so speed depends on it.

Still stuck?

Email info@quantonique.com with the column headings from your file — the first row only, never the data itself — and we will tell you how to map it.

Open the app