RiskBoard
Demonstration — sample data (Helvetia Métal SA, industrial SME). Read-only.Create my workspace to enter your own risks →

From your file to the report

A fictitious risk register, kept in Excel the way many SMEs keep it, flaws included. Everything below is calculated by the import screen's code from this file, with no manual correction.

10 rows below the header, “;” separator, UTF-8: it opens in Excel like the register export. Company, roles and figures are invented.

Download the fictitious example (CSV)

The import screen exists in French only, and it recognises columns from French headers. The example file is therefore in French too; a column with an English header is assigned by hand, from the list.

How the file gets in

There is no file upload. You open the file in Excel, select the rows including the headers, copy them and paste them under “Registre”, “Importer depuis Excel”. Nothing is saved before you click “Importer”, and 300 rows at most go through per paste.

1. The columns taken over

The screen guesses the mapping from the file's headers; each choice can be corrected from a list. Only the risk title is required.

File columnBecomes
N°Ignored
RisqueRisk title
CatégorieCategory
DescriptionDescription
PropriétaireOwner
ProbabilitéNet likelihood
ImpactNet impact
Mesures en placeControls in place
Plan d'actionAction plan
Responsable actionAction owner
ÉchéanceAction due date
RemarqueIgnored

2. What the file lacks

No column fills these fields. The import still goes ahead, with these consequences:

The columns “N°” and “Remarque” match no field: their content is not imported. Risks receive the codes R-01, R-02… in file order.

3. What the screen flags before importing

These warnings appear below the preview, before you click “Importer”:

Empty rating, replaced by 3: 1 of 9 (row 5 of the file).

Due date that does not exist in the calendar, left empty: 1 of 9 (“31.09.2026”).

Row without a risk title, ignored: 1.

No warning, because the conversion is unambiguous: ratings written in words become numbers (“Élevé” → 4, “Moyen” → 3). The preview shows the original value under each rating.

What the screen does not flag

The duplicate. Rows 2 and 7 of the file hold the same risk. The import does not compare rows with each other: they become two risks, and one of them has to be deleted by hand in the register.

4. The resulting register

The risks as they arrive in the register, ranked by net level (likelihood × impact), with no retouching.

Risks imported
9
Above appetite (net ≥ 12)
5
Default ratings
1
Duplicates imported
1

Of the 5 risks above appetite, 1 comes from the duplicate and 1 rests on a default likelihood: the board would be reading wrong figures.

CodeRiskNet
R-01
Cyberattaque par rançongiciel
Informatique
20 · Critical
R-06
Cyberattaque par rançongiciel
Informatique
20 · Critical
R-04
Défaillance du principal client
Commercial
15 · Critical
R-02
Départ du chef d'atelier
Ressources humaines
12 · High
R-03
Hausse du prix de l'électricité
Financier
12 · High
R-07
Panne de la presse principale
Opérationnel
9 · Medium
R-08
Retard de livraison de l'acier
Approvisionnement
9 · Medium
R-05
Accident sur une presse
Santé-sécurité
8 · Medium
R-09
Données personnelles mal protégées
Juridique
6 · Medium

5. What the file does not hold: the settings

The report needs settings that an Excel register does not carry. They are entered once under “Paramètres”:

6. To check before presenting the report

This work remains manual, risk by risk:

  1. Delete the duplicate (rows 2 and 7 of the file).
  2. Rate the likelihood left at 3 (row 5 of the file).
  3. Enter gross ratings where the controls in place reduce the risk.
  4. Correct the unreadable due date (“31.09.2026”).
  5. Review the appetite threshold and the period under “Paramètres”.

The demonstration report comes from another fictitious register, Helvetia Métal SA's, which runs through the whole demonstration: this is what the deliverable looks like once the register has been checked.

See a complete board report →