# Import & Export Bulk import and export of Inventory assets, using the same preview-before-confirm workflow used elsewhere in the portal. > **Required permission:** `Inventory Import` for import, `Inventory Export` for export — separate from the base `Inventory` permission. --- ## Import workflow 1. Navigate to **Import** → **Inventory** tab 2. Upload your CSV file 3. Click **Preview** 4. Review the preview, then **Confirm** ### Step 2 — Preview Before anything is written, the portal shows: - **New catalogue entries** — any category, sub-category, or vendor name in the file that doesn't exist yet is listed explicitly, so a typo ("Sim" vs "SIm") is caught before it becomes two near-duplicate categories, not months later while filtering the list. Names that only differ by case count as the *same* entry. - **Row-by-row result** — each row marked **New**, **Update**, or **Error**, with the reason for any error spelled out (e.g. "Category is required.") ### Deduplication A row matches an existing asset by: 1. **Serial Number**, if the row has one, or 2. **Category + Name + Assigned User UPN**, otherwise A matching row updates the existing asset instead of creating a duplicate — this is also what makes the exported file safe to re-import unchanged (see below). > Assignment fields (assigned user, device) on an existing asset are only changed if the CSV row actually provides a value for them — a re-import that doesn't mention the assigned user never clears an assignment set through the portal. > **Empty cells clear the other fields.** When a row updates an existing asset, an empty cell empties the corresponding field (for example the expiry date or the cost), and an empty `Stato` sets the asset back to **Active**. To change only some fields, start from an [export](#export) so that the file contains the current values of all the others. --- ## CSV format | Property | Value | |----------|-------| | **Format** | CSV, semicolon-separated (`;`) | | **Encoding** | UTF-8 | | **Header row** | Required (first row) | | **Date format** | `yyyy-MM-dd` | | **Decimal format** | Point as decimal separator (`12.50`, never `12,50`) | | **Maximum rows** | 5,000 per file | Download a ready-to-use template from the portal: **Import → Inventory → Download Template**. ### Columns ```csv Categoria;SottoCategoria;Nome;Vendor;Seriale;OrdineAcquisto;Stato;Costo;PeriodicitaCosto;Scadenza;DataAssegnazione;UtenteUPN;Note Software;Licenze;Microsoft 365 E3;Microsoft;;PO-2026-0001;Active;12.50;Monthly;2027-06-30;2026-01-15;mario.rossi@cliente.it;Rinnovo automatico SIM;;SIM aziendale 3934;Vodafone;8939000000000000001;;Active;9.90;Monthly;;;; ``` | Column | Required | Type | Description | |--------|----------|------|-------------| | `Categoria` | **Yes** | Text | Category name — created in the catalogue if new | | `SottoCategoria` | No | Text | Sub-category name — created in the catalogue if new | | `Nome` | **Yes** | Text (max 200) | Asset display name | | `Vendor` | No | Text (max 150) | Vendor/publisher name — created in the catalogue if new | | `Seriale` | No | Text (max 150) | Serial number, licence key, or ICCID — used for deduplication when present | | `OrdineAcquisto` | No | Text (max 150) | Purchase order reference | | `Stato` | No | Fixed value | See **Status values** below. Defaults to `Active` if left empty | | `Costo` | No | Decimal | Invariant culture, point as decimal separator | | `PeriodicitaCosto` | No | Fixed value | See **Cost period values** below | | `Scadenza` | No | Date (`yyyy-MM-dd`) | Expiry or renewal date | | `DataAssegnazione` | No | Date (`yyyy-MM-dd`) | Assignment date | | `UtenteUPN` | No | Text (max 255) | Assigned Entra user (UPN) | | `Note` | No | Text | Free text | ### Status values — always accepted in English, Italian aliases also accepted on import | English (used by export) | Italian alias (accepted on import) | Meaning | |---------------------------|--------------------------------------|---------| | `Active` | `Attivo` | In active use | | `Parked` | `Parcheggiato` | Temporarily set aside | | `Dismissed` | `Dismesso` | Retired / no longer in use | ### Cost period values — always accepted in English, Italian aliases also accepted on import | English (used by export) | Italian alias (accepted on import) | |---------------------------|--------------------------------------| | `OneTime` | `Una tantum` / `UnaTantum` | | `Monthly` | `Mensile` | | `Yearly` | `Annuale` | > **On export**, `Stato` and `PeriodicitaCosto` are always written using the English invariant names above, regardless of the portal's display language — this is what makes an exported file safe to re-import into any tenant, in any language, without editing it first. --- ## Export 1. Navigate to **Export** from the sidebar 2. Select **Inventory** 3. Click **Export Data** The exported file uses **exactly the same columns, in the same order**, as the import format above — category, sub-category, and vendor are written as their catalogue names (never as internal IDs), dates as `yyyy-MM-dd`, cost with a point as decimal separator, and `Stato`/`PeriodicitaCosto` in English. ### Round-trip Because the export format matches the import format column-for-column, and every asset already has a serial number or a category+name+user combination that the import can match, **an exported file can be re-imported without any changes** and every row is recognized as an update to the asset it came from — no duplicates, no new catalogue entries created. --- ## Tips to avoid errors - **Use the template.** It guarantees the correct column names, order, and separator. - **Watch for near-duplicate category names.** The preview lists every new catalogue entry the import would create — check that list before confirming, especially for typos like "Sim" vs "SIm" (these two collapse into one entry automatically, since matching is case-insensitive). - **Use a point, not a comma, for `Costo`.** A value like `12,50` is rejected rather than silently read as `1250`. - **Keep dates in `yyyy-MM-dd`.** Any other format is rejected for that row. - **Review the preview.** Rows with errors are listed with the specific reason and are skipped — fix the CSV and re-upload to include them.