Skip to main content

​
Overview

The current plot import in Polygon One is a staged review workflow. You do not simply upload a file; instead, before the final import you see a structured preview with notes on precision, country plausibility, and individual problematic rows. You usually open this workflow via Import plots. The supplier portal uses the same technical core, but there with stricter restrictions for suppliers.

​
Start from the template

Before you build a file of your own, pick up the scaffold via Download template:
  • Download Excel template (recommended) — an .xlsx with an Import sheet and a second Guide sheet naming each column’s requirement, description and accepted values.
  • Download CSV template — the same columns as a text file.
The template carries the columns Plot Name, Geometry, Latitude, Longitude, Commodity Group, Production Start Date, Production End Date, Common Name, Scientific Name, Country, Producer Name and Size (ha). In both of its sample rows the geometry sits as WKT in the Geometry column, once as a point and once as a polygon; Latitude and Longitude stay empty there. The two coordinate columns are the alternative when you have no WKT geometry.

​
Supported File Formats

  • GeoJSON and .json
  • CSV
  • Excel (.xlsx, .xls)
Depending on the file, Polygon One recognizes, among other things:
  • GeoJSON geometries
  • WKT geometry columns (point, multipoint, polygon, multipolygon)
  • Point lists with latitude and longitude
  • Properties such as area name, country, producer, and production data
Shapefiles are not read. Open them in QGIS and export as GeoJSON first — the upload dialog says so too.Coordinates are expected as decimal degrees only. Degrees/minutes/seconds are not recognized.

​
Decimal commas

In a semicolon-delimited CSV, latitude, longitude and the area may carry a decimal comma: -15,654321 is read correctly. Ambiguous cells such as 1,234 — decimal separator or thousands separator? — are deliberately left alone and surface as an invalid coordinate rather than being silently turned into the wrong number. This does not apply to comma-delimited CSV files or to Excel files. Use a decimal point there. This holds for coordinates only: an unreadable Size (ha) does not announce itself, it simply stays empty — and only surfaces later, when a point plot without an area cannot be imported.

​
Dates

Production start and production end are read to the day, as DD.MM.YYYY (always day first) or YYYY-MM-DD — so 01.03.2026 or 2026-03-01. A trailing clock time of the kind ERP exports write (01.03.2026 00:00:00) is discarded, and a two-digit year is read the way Excel reads it (00–29 as 2000s, 30–99 as 1900s). This holds for tabular files and for GeoJSON properties alike.
Slashed dates are refused, not guessed. 03/01/2026 can be 3 January or 1 March; Polygon One does not decide that for you. The row lands in the issue register as an error row instead — the production period is not silently dropped. An ISO value with a time is read too (2026-03-01T00:00:00, with or without a trailing Z); only a written-out time-zone offset such as +02:00 is refused, because a production date is a calendar day. Convert the date columns of your source file to one of the two accepted spellings before uploading.

​
The Process in Detail

1

Assign supplier and articles

First, assign the import to a supplier and the affected articles. This assignment controls which supplier-article combination the areas are later available for in compliance, the map, and the DDS.
2

Upload the file

Upload the file. Polygon One accepts GeoJSON, CSV, and Excel and first processes the upload into a prepared preview rather than directly into the production inventory.
3

Review detected columns and mapping

The system largely detects the file format and columns automatically. Via Adjust column mapping you correct it when the file uses unusual headers — for:
  • Plot name
  • Country
  • Producer name
  • Production start date and production end date
  • Area / size for point geometries
  • Geometry column or Latitude column and Longitude column — either/or, never both
  • Coordinate order
If detection finds no coordinates at all, the file is not rejected. Instead the column picker opens with the note “We could not tell which columns of this file hold the coordinates. Nothing is wrong with the file — pick the coordinate columns and we will read it again.” You pick the two columns; the already-uploaded file is read again without a re-upload.
4

Read the review panel and make decisions

After staging, you see a review interface with:
  • Country plausibility check for coordinates outside the expected production country
  • Precision histogram for the coordinates
  • Normalization warnings, such as automatically corrected coordinate order or removed polygon holes
  • Issue register with error rows and precision rows
  • Search, tabs, and export for the issue list
5

Confirm the import

Before the final import, you see a summary of how many areas:
  • are imported cleanly and directly
  • are imported with confirmed low precision
  • are deliberately skipped

​
Row-Level Decisions

For flagged rows, the following decisions are available depending on the data situation:
Low precision does not automatically mean incorrect geodata. The preview deliberately distinguishes between technical plausibility and format/precision questions.

​
Bulk Actions and Issue List

For larger files, the workflow offers additional bulk actions:
  • Confirm & pad all flagged
  • Skip low-precision plots under 2 ha
  • Reset decisions
  • Export issue list
The export is especially helpful when you want to clean up problematic rows with your supplier or data provider outside the platform.

​
Rows that are never imported at all

Some rows are refused when you confirm the import and are then listed under Rows not imported — each with its own reason:

​
Re-importing the Same Plot

If you import a plot that already exists — same name, same country, same geometry, same article links — the row is not discarded. Corrected values are applied:
  • Producer name, production start, production end, and species (scientific name) are updated when the file supplies a value.
  • A field left empty in the file overwrites nothing — an existing value is kept.
  • Geometry, country, name, and article links are untouched. This is a correction of the declaration, not a new outline.

​
What you end up with

The closing message names four counters separately — “Created: …”, “Updated: …”, “Left unchanged: …” and “Not imported: …”. A re-import with no substantive change writes nothing and counts as left unchanged; that is not an error. The heading sums the run up: Import complete when no row was refused, Import partially complete when some were, and Import not completed when every attempted row was refused. When nothing was refused and nothing was created or updated either, it reads No changes made. Created and updated plots stay saved either way — the run is not rolled back. Rows you skipped in the review panel count as left unchanged — they do not stop an Import complete. If rows were refused, they are listed below under Rows not imported, each with its reason; a long list ends with ”… and … more rows”.
This is the simplest way to correct a wrong production period or a missing producer name across many plots: fix the file, import again, check the “Updated” counter. Which plots were affected is recorded in the audit log under “Updated plots”.

​
After the Import

After a successful import, you manage the areas in two places:

Plots

Table view with filters, precision check, bulk actions, and exports.

Map

Geographic control, detailed review, and risk assessment on the map.
If a file contains many problematic rows, first export the issue list, clean up the source outside of Polygon One, and re-upload the corrected batch. This is usually faster than a fully manual row-by-row reconciliation.