This page describes how to configure Pivot table, CFP Pivot table, and Dynamic table elements in a form template in Lucanet Data Collection — adding and configuring columns and rows, and setting up drill down. For instructions on how to create a form template and add elements to it, see Creating and Configuring Form Templates.

Configuration options in Data Collection vary depending on the table type:

  • Pivot table and CFP Pivot table support the full range of column types, row configuration, and drill down described on this page.
  • Dynamic table supports all column types. Rows are added by collectors during data collection, so a dynamic table has no row configuration in the template. Drill down is not available for dynamic tables.

When you add a CFP Pivot table element to a form template, Data Collection asks you to choose which rows to include from the selected Consolidation & Financial Planning (CFP) workspace before the table is created. This row selection step lets you control exactly which eligible rows end up in the new table, instead of pulling all of them in automatically.

To add a CFP-based pivot table to a form template in Data Collection:

1

Add a CFP Pivot table element to a section (see Adding Elements). The Add Element dialog is displayed:

Shows the 'Add Element' dialog in Data Collection with the text 'Select the ledgers or schedules from Consolidation & Financial Planning to be imported.', the 'Select ledger or schedule' drop-down list, and the 'Cancel' and 'Import' buttons.
'Add Element' dialog
2

Select the Consolidation & Financial Planning workspace whose structure you want to import and click Import. The Select Rows to Include dialog is displayed, showing all eligible rows of the selected workspace in their original hierarchy:

  • For General ledger, Subledger, or Statistical Ledger workspaces, only account hierarchy rows are displayed. Formulas, separators, and total lines are excluded.
  • For Schedule workspaces, all workspace structure rows are displayed. Formulas and total lines are excluded.
Shows the 'Select Rows to Include' dialog in Data Collection with the text 'Choose which eligible rows of Bug rounding are included in the new pivot table.', the '0 rows selected' counter, the 'Select all', 'Deselect all', 'Expand All', and 'Collapse All' options, the 'Show items only' toggle, the 'Search rows' field, the rows 'A' to 'F' with check boxes, the message 'Superordinate rows will be moved along with the selected subordinate rows.', and the 'Back' and 'Confirm' buttons.
'Select Rows to Include' dialog
3

Choose one of the available selection options:

  • Select all — selects all eligible rows across all levels.
  • Show items only — selects only rows of type Item in the CFP account hierarchy, regardless of their position in the hierarchy.
  • Custom selection — lets you check or uncheck individual rows. Selecting a superordinate row automatically selects all of its descendants (unless Show items only is active).
4

Click Confirm. The CFP Pivot table is created with the selected rows, preserving their original hierarchy.

For Schedule workspaces, columns are also pre-populated from the workspace. For General ledger, Subledger, and Statistical Ledger workspaces, the table starts without columns — add them as described in Adding Columns.

A selected superordinate row whose subordinate rows were not selected is included as a standalone row. If only some subordinate rows of a superordinate row are selected, the superordinate row travels with the selected subordinate rows to preserve the hierarchical structure.

When a CFP pivot table is created, Data Collection records a snapshot of the structure of the source workspace in CFP. Because the workspace can change in CFP after the table was created — rows can be added, renamed, moved, or deleted — you can check an existing table for structural changes and adjust its row selection.

To check a CFP pivot table for structural changes in Data Collection:

1

Click the Sync with CFP icon in the CFP pivot table in the detail view of the form template. Data Collection compares the recorded snapshot with the current CFP workspace and displays the Select Rows to Include dialog.

2

Review the changes. A message at the top of the dialog summarizes the number of detected changes per type, and each changed element is marked with a colored label:

  • Added — the element exists in CFP but is not part of the table.
  • Deleted — the element is part of the table but no longer exists in CFP.
  • Updated — the element's name, superordinate element, or order has changed in CFP.
Shows the 'Select Rows to Include' dialog in Data Collection with the message '164 changes detected in the CFP workspace since this table was configured: 164 added, 0 renamed, 0 deleted.', the '9 rows selected' counter, selected rows of the 'Bilanz' workspace such as '3. goodwill' and '4. prepayments for intangible assets', unselected rows labeled 'Added' such as 'ii. tangible fixed assets', and the 'Back' and 'Confirm' buttons.
Detected CFP changes in the 'Select Rows to Include' dialog
3

Adjust the row selection — for example, select added elements to include them in the table, or deselect elements that no longer exist in CFP — and click Save. The snapshot is updated to the saved state.

Additional notes:

  • The Sync with CFP icon is available to Data Collection administrators only. You can also use the icon to reconfigure the row selection of an existing table when no structural changes are pending.
  • Data that has already been collected for changed rows is preserved.
  • If the CFP workspace no longer exists in CFP, the table and its collected data are not affected.
  • When you duplicate a CFP pivot table — directly or by duplicating its section or form template — a new snapshot is recorded for the copy.

In Data Collection, columns define the data points you want to collect for each row in the table. Click Add column in the table header and select one of the available column types.

The available column types depend on the table type.

Creates a column from scratch where you define all properties manually.

Shows the 'Add Column' dialog in Data Collection with the 'Name' field containing 'Test', the 'Type' drop-down list set to 'Currency', the 'Period value type' drop-down list set to 'End of period', the 'Decimal places' drop-down list set to '2', the 'Entry method' drop-down list set to 'Manual & import (default)' with the description 'Users type values in the UI; imports also write to this column.', the 'Used for postings' and 'This column is mandatory' check boxes, and the 'Cancel' and 'Add' buttons.
'Add Column' dialog

The following table describes the fields for an empty column in Data Collection:

FieldDescription
NameUnique name for the column within the table.
TypeData type for the column values. Available types: Numeric (float), Numeric (integer), Currency, Text.
Period value typeDefines which type of period data is retrieved when importing from Consolidation & Financial Planning. Available for Numeric and Currency columns. See Period Value Type for CFP Import for details.
Used for postingsWhen activated, this column is included in the posting data that CFP imports from Data Collection. Available only for Currency columns in CFP Pivot table elements. One or more currency columns per table can be activated. Postings are supported for CFP pivot tables of the types Balance Sheet, P&L, Schedule, and Statistical Ledger. For the full import workflow, see Importing Data Collection Postings into CFP.
Decimal placesSpecifies the number of decimal places. Available for Numeric (float) and Currency column types.
Entry methodControls how data enters the column. Available for editable columns in pivot tables, CFP pivot tables, and dynamic tables. Available options:
• Manual & import — the default; data can be entered manually or via any import channel.
• Manual only — the column accepts manual entry only; Excel, Lucanet.Financial Warehouse, and CFP imports skip these columns.
• Import only — collectors cannot edit the column during data collection; it receives values through imports only.
Manual only and Import only columns are marked with dedicated icons in the column header. The setting is captured when a data collection process starts — later changes apply only to new processes.
Text typeDefines whether the text field allows single-line or multi-line input. Available for Text column type.
This column is mandatoryWhen activated, data collectors must provide a value for this column.

Period value type is a required setting for Numeric and Currency columns in Pivot tables and CFP Pivot tables. It defines which type of period data Data Collection retrieves when importing from Consolidation & Financial Planning, so that each column gets the right values — for example, opening balances, within-period transactions, or closing balances.

The following table describes the available period value types in Data Collection:

Period value typeDescription
Beginning of periodValue at period start
Within periodChanges during the period
End of periodValue at period end
YTD valueCumulative from fiscal year start
Standard (auto)Lets Consolidation & Financial Planning determine the value based on account type.
N/AExcludes the column from the CFP import, so the column keeps its content.

Period value type is configured when creating the column. Once data has been collected in the column, this setting cannot be changed.

Creates a column based on dimension elements from Consolidation & Financial Planning. The following dimensions are available in Data Collection:

  • Transaction types
  • Partners
  • Adjustment levels
Shows the 'Add Column from CFP' dialog in Data Collection with the text 'Select the dimensions from Consolidation & Financial Planning to be used as columns.', the dimensions 'Transaction types', 'Partners', and 'Adjustment levels', and the 'Cancel' button.
'Add Column from CFP' dialog

CFP pivot tables based on Schedule workspaces automatically include Value query columns. Value query is a transaction type specific to Schedule workspaces in CFP. These columns appear alongside any other transaction-type columns in the table.

Value query columns in Data Collection behave as follows:

  • Read-only for collectors — the values are calculated from the CFP schedule source and cannot be edited during data collection.
  • Available as operands — you can reference value query columns in total and formula column expressions.
  • Included in CFP import — values are included when collectors import data from CFP into the form.
  • Excluded from Excel and Lucanet.Financial Warehouse import — Value query columns are skipped when importing from those sources.

Creates an automatically calculated column that aggregates values from other columns.

Shows the 'Add Total Column' dialog in Data Collection with the 'Name' field containing 'Total', the 'Type' drop-down list set to 'Currency', the 'Decimal Places' drop-down list set to '2', the empty 'Select columns' drop-down list with the error message 'This field is required.', and the 'Cancel' and 'Add' buttons.
'Add Total Column' dialog

The following table describes the fields for a total column in Data Collection:

FieldDescription
NameUnique name for the column within the table
TypeData type for the aggregated values. Available types: Numeric (float), Numeric (integer), Currency.
Decimal placesNumber of decimal places for the result. Available when Currency or Numeric (float) is selected.
Select columnsThe columns whose values are aggregated.

The total column is only available when your table contains at least two numeric or currency columns. Calculated columns (such as existing total columns) cannot be included in a new total column.

Creates a column whose values are calculated automatically from an arithmetic expression. Only administrators can add formula columns. Formula columns display the ∫ icon in the column header alongside a lock indicator.

Shows the 'Add Column' dialog for a formula column in Data Collection with the 'Name' field containing 'Test', the 'Data type' drop-down list set to 'Numeric (integer)', the 'Formula' field containing the operand 'Ibidem col 5.Sec.Table 2.Col 1', the 'Decimal places' drop-down list set to '2', and the 'Cancel' and 'Add' buttons.
'Add Column' dialog for a formula column

The following table describes the fields for a formula column in Data Collection:

FieldDescription
NameUnique name for the column within the table
Data typeData type for the calculated values: Numeric or Currency. Determines which same-table columns are eligible as operands.
FormulaThe arithmetic expression, built from operand columns and numeric constants
Decimal placesNumber of decimal places for the result

Operand scope

A formula column expression in Data Collection can reference:

  • Columns in the same table that match the selected data type
  • A column from another table within the same form template. You also specify how the source row is resolved:
    • Fixed row — uses one specific row from the source table for every row in the formula column. Useful when the source table contains a single reference value such as an exchange rate.
    • Matched row — uses the row from the source table with the same name as the current row. Useful when both tables share the same row structure.
  • A column from a table in a different form template — resolved at collection time for the same reporting entity and period
  • Numeric constants — for fixed rates or divisors, for example × 0.25 for a 25% tax charge

Key behaviors

The following key behaviors apply to formula columns in Data Collection:

  • Collectors can hover over a formula cell to see its expression.
  • An empty operand cell is treated as 0 when the formula is evaluated, so the result is always calculated. The empty cell itself remains empty. This also applies to operands from other tables and form templates, even if the referenced form has not yet been submitted.
  • The formula is calculated at every level of a row hierarchy, including superordinate rows and the Total row.
  • Division by zero produces an empty result. This also applies when the divisor is an empty cell.
  • A formula cell remains empty if the table contains no data rows or if, with Matched row, no row with the same name exists in the source table.
  • References are stored by internal ID. Renaming a row or column never breaks a formula.
  • Formula columns can be used as operands in intercompany validation rule definitions.

If the table contains no numeric or currency columns when you create the formula column, the formula builder indicates that no same-table columns are currently available. Operands from other tables remain accessible, and same-table columns can be added later.

Row configuration is available only for Pivot table and CFP Pivot table elements. Dynamic table rows are added by collectors during data collection.

In Data Collection, rows define the items or categories for which you want to collect data. Click Add row at the bottom of the table and select one of the available row types.

Creates a new row where you define the name manually.

Shows the 'Add row' drop-down menu below a table in Data Collection with the options 'Add empty row', 'Add formula row', and 'Add total row'. Highlighted in red is the 'Add empty row' option.
'Add empty row' option

Creates a row that automatically aggregates values from other rows.

Shows the 'Add row' drop-down menu below a table in Data Collection with the options 'Add empty row', 'Add formula row', and 'Add total row'. Highlighted in red is the 'Add total row' option.
'Add total row' option

Creates a row whose values are calculated automatically from an arithmetic expression. Only administrators can add formula rows. Formula rows display the ∫ icon next to the row name.

To add a formula row, click + Add row and select Add formula row.

Shows the 'Add formula row' dialog in Data Collection with the 'Name' field containing 'Test', the 'Formula' field containing the operands 'ft-11may.section1.lbl1(1).Profit.test' and 'Ibidem col 5.Sec.Table 2.Row 1' joined by '*', and the 'Cancel' and 'Add' buttons.
'Add formula row' dialog

The following table describes the fields for a formula row in Data Collection:

FieldDescription
NameUnique name for the row within the table
FormulaThe arithmetic expression, built from operand rows and numeric constants

Operand scope

A formula row expression in Data Collection can reference:

  • Rows in the same table — including regular data rows, the total row, and other formula rows
  • A row from another table within the same form template. You also specify how the source column is resolved:
    • Fixed column — uses one specific column from the source table for every column in the formula row. Useful when the source table contains a single reference value such as an exchange rate.
    • Matched column — uses the column from the source table with the same name as the current column. Useful when both tables share the same column structure.
  • A row from a table in a different form template — resolved at collection time for the same reporting entity and period
  • Numeric constants

Key behaviors

The following key behaviors apply to formula rows in Data Collection:

  • Collectors can hover over a formula cell to see its expression.
  • An empty operand cell is treated as 0 when the formula is evaluated, so the result is always calculated. The empty cell itself remains empty. This also applies to operands from other tables and form templates, even if the referenced form has not yet been submitted.
  • Division by zero produces an empty result. This also applies when the divisor is an empty cell.
  • A formula cell remains empty if the table contains no data rows or if, with Matched column, no column with the same name exists in the source table.
  • References are stored by internal ID. Renaming a row or column never breaks a formula.
  • Formula rows cannot be used as operands in intercompany validation rule definitions.
  • A formula row can be placed at any position in the table except after the Total row. If one formula row references another, Data Collection calculates them in the correct sequence automatically.

The Total row always excludes formula rows from its sum to prevent double-counting of derived values.

To add a subordinate row under an existing row, right-click the row and select Add row. The submenu offers the following options in Data Collection:

OptionDescription
EmptyAdds an empty subordinate row under the selected row.
Based on the CFP Partner dimensionOpens the Add row based on CFP dialog to add subordinate rows based on the Partner dimension.
Shows the context menu of a row in Data Collection with the options 'Add row', 'Drill down', and 'Delete', and the 'Add row' submenu with the options 'Empty' and 'Based on the CFP Partner dimension'.
Adding a subordinate row

The Add row option is not available on total rows.

Drill down is available only for Pivot table and CFP Pivot table elements.

In Data Collection, drill down allows data collectors to provide more detailed data at a lower, reporting entity-specific level. You configure a drill down for rows, for columns, or for both in the same table. The following dimensions are available:

  • Local account
  • Cost center
  • Partner
  • Transaction type
  • Adjustment level

Each dimension can be used either for rows or for columns, not for both. A dimension that is already used for a row drill down is disabled in the column drill down configuration, and vice versa. The number of drill down levels for rows or columns is limited only by the number of dimensions not yet used for the other one.

For CFP pivot tables, the CFP import transfers data into the configured drill down automatically. For more information, see Importing Data from Consolidation & Financial Planning.

To configure a drill down for a row in Data Collection:

1

Right-click the row.

2

Select Drill down from the context menu. The Configure Row Drill Down dialog is displayed:

Shows the 'Configure Row Drill Down' dialog in Data Collection with the text 'Configure your drill down for Positionskonto - Non-current assets.', the 'Drill down by' check boxes 'Local account', 'Cost center', 'Partner', 'Adjustment level', and 'Transaction type', the level labels 'Level 1', 'Level 2', and 'Level 3' next to the activated 'Local account', 'Cost center', and 'Partner' check boxes, the 'Require transaction currency details' toggle with its description, and the 'Cancel' and 'Save' buttons.
'Configure Row Drill Down' dialog
3

Under Drill down by, activate the check boxes of the dimensions in the order of the drill down levels. Each activated dimension receives a level label, e.g. Level 1 for the first dimension you activate. Dimensions already used for a column drill down are disabled.

4

Optionally, if Partner is one of the selected dimensions, activate the Require transaction currency details option. Collectors must then enter a transaction currency and the value in transaction currency for each partner in the drill down dialog.

5

Click Save.

For each drill down level, the row displays a drill down icon Drill down icon. Point at an icon to display the dimension of that level.

Rows with subordinate rows

You can also configure a row drill down on a row that has subordinate rows. The configuration applies to all lowest-level rows beneath the selected row, including rows you add later, and replaces any row drill down configured directly on these rows. In the table, rows with an inherited configuration display grey icons, rows with their own configuration display purple icons.

Shows a CFP pivot table in the template builder in Data Collection with purple drill down icons on the rows '1. revenue', '2. increase o...', and 'Test row' and grey drill down icons on their lowest-level rows such as 'revenue' and 'Test Child 1', the 'Import drill down structure' button, and the message 'No postings will be generated for this table unless a currency column is configured.'
Drill down icons for own and inherited row drill down configurations

If you remove the configuration of a superordinate row with Delete drill down configuration, the drill down is removed from all its lowest-level rows. Configurations that were replaced when you configured the superordinate row must be configured again.

During data collection, collectors add the elements of the selected dimensions as subordinate rows directly in the table, one level per dimension. For more information, see Entering, Validating, and Approving Data.

You can configure a drill down for columns with Numeric or Currency data types. To configure a drill down for a column in Data Collection:

1

Click the three-dot icon in the column header.

2

Select Drill down from the menu. The Configure Column Drill Down dialog is displayed:

Shows the 'Configure Column Drill Down' dialog in Data Collection with the text 'Configure your drill down for 2025.' and the 'Drill down by' check boxes, where 'Local account', 'Cost center', and 'Partner' are disabled with the explanation 'Already used for rows, unavailable for column drill down', 'Adjustment level' and 'Transaction type' are available, and the 'Cancel' and 'Save' buttons.
'Configure Column Drill Down' dialog in a table with a row drill down
3

Under Drill down by, activate the check boxes of the dimensions in the order of the drill down levels. Dimensions already used for a row drill down are disabled.

4

Click Save.

When a column drill down is configured, data collectors enter the values of that column for each element of the selected dimensions in the drill down dialog, across all rows in the table.

When a table has both a row drill down and a column drill down, collectors enter each cell at the intersection of a row element and the column per element of the column dimensions in the drill down dialog. Collectors cannot type values directly into these cells.

If all five dimensions are used for the row drill down, a column drill down is not available in that table.

This section applies to Pivot table and CFP Pivot table elements. Reordering is only available while the form template is in Draft state.

You can change the display order of rows and columns in a pivot table or CFP pivot table in Data Collection by drag-and-drop. Rows can also be moved under a different superordinate row within the same table. Both operations only change the display structure — collected data, drill down configurations, and aggregations are bound to row and column identifiers, not to their position, so no data is lost or recalculated unexpectedly.

You can drag any row to a new position among the rows at the same level (rows under the same superordinate row). The Total row is fixed at the bottom of the table and cannot be moved.

When you drag a superordinate row, all its subordinate rows move with it as a single unit. The internal superordinate–subordinate relationships within the moved subtree are preserved.

A horizontal line indicates the drop position during the drag. If you drop a row on an invalid target, it returns to its original position and an error message is displayed.

You can drop a row onto another row to make it subordinate to that target row, changing its position in the hierarchy.

Shows rows of a table in Data Collection under the superordinate row 'Revenue', such as 'testa', 'Positionskonto - Revenue', '3000 Revenue (third)', and '3001 Revenue (foreign)', while the row '3070 Other operating income' is dragged onto another row.
Moving a row to another superordinate row

When you drop a row on a valid target, a Move Row dialog is displayed showing the row you are moving, the source superordinate row, and the target superordinate row:

Shows the 'Move Row' dialog in Data Collection with the question 'Move 3070 Other operating income from Revenue under 5100 Other operating income?' and the 'Cancel' and 'Move' buttons.
'Move Row' dialog

Review the details and click Move to commit the move.

After the move, the following applies in Data Collection:

  • The aggregation of the source superordinate row decreases by the moved row's contribution.
  • The aggregation of the target superordinate row increases by the same amount.
  • The Total is unaffected — it aggregates the values of all the lowest-level rows regardless of hierarchy.
  • Drill down configurations travel with the moved row and continue to work.

The following moves are not allowed in Data Collection:

RestrictionReason
A row cannot be moved under one of its own subordinate rows.Circular hierarchy.
A row cannot be moved under the Total row.The Total row is fixed.
A manually-added CFP partner-based row cannot be moved under another CFP partner-based row.CFP partner-based rows are never valid targets.
A row cannot be moved under a lowest-level row that already has a drill down configured.Rows with a drill down cannot accept subordinate rows.
A row cannot be moved beyond the maximum supported hierarchy depth.Depth limit.

You can drag any column to a new position within the table. All column types are draggable.

The following exceptions apply when reordering columns in Data Collection:

  • The first column of the table is fixed.
  • The last action column is fixed.

When you move a Total column, it continues to reference its source columns by identifier. Moving the total column or its source columns has no effect on the calculated value.

Rows and columns can only be moved within their own table — moving them between tables is not supported.

This content was generated using AI and reviewed by Lucanet subject matter experts before publication.