Transaction data with a time element (orders, invoices, stock, forecasts, actuals) is viewed, assessed and analysed as a pivot over the golden dimensions: any hierarchy level on the rows, months, weeks or any dimension on the columns, one or several measures in the cells, calculated measures beside the stored ones, a transposed view with one click.
| node | Jul actual | Jul plan | Δ | Aug actual | Aug plan | Δ | Sep actual | Sep plan | Δ |
|---|---|---|---|---|---|---|---|---|---|
| Power supplies | 48,200 | 46,000 | +4.8% | 51,900 | 48,000 | +8.1% | 44,100 | 50,000 | −11.8% |
| Modular 600W | 21,400 | 20,000 | +7.0% | 23,100 | 21,000 | +10.0% | 18,200 | 22,000 | −17.3% |
| Modular 850W | 15,800 | 15,500 | +1.9% | 16,900 | 16,000 | +5.6% | 15,100 | 16,500 | −8.5% |
| Fixed 450W | 11,000 | 10,500 | +4.8% | 11,900 | 11,000 | +8.2% | 10,800 | 11,500 | −6.1% |
| Controllers | 9,400 | 9,000 | +4.4% | 9,100 | 9,500 | −4.2% | 9,900 | 9,800 | +1.0% |
| Total | 57,600 | 55,000 | +4.7% | 61,000 | 57,500 | +6.1% | 54,000 | 59,800 | −9.7% |
Rows are the product hierarchy at family and group level; columns are months; each month shows three measures: actual, plan and the variance, which is a calculated measure. Switch the rows to plant or customer with a chip; switch the columns to weeks; transpose to put time down the rows; click a cell to see the order lines behind it.
Rows and columns from any two dimensions; nest a second dimension on the rows (family, then plant).
Several measures per cell, or measures as columns: actual, plan, variance, share of parent.
Two measures from two systems for the same golden record, side by side with the difference explained.
Read the page →Time down the rows, measures across the columns: the shape finance reads.
Formulas over other measures, saved once, reused everywhere; unit-aware.
Read the page →From any cell to the transaction rows and to the source record.
Read the page →A pivot needs the hierarchy, the calendar and the transactions in one place with keys between them. Master-data products usually hold the hierarchy alone and leave the facts to a warehouse, and planning tools hold the facts but lack the golden record's matching, survivorship and cross-reference. Lima's bed holds both, so the pivot is one query and the roll-up follows the golden hierarchy.
A merge changes the pivot at once: the absorbed customer's orders roll up under the surviving one, in every version of every hierarchy, with no reload.
As big as the engine can aggregate: the showcase pivot over 214,000 order lines and 9,340 suppliers answers in about two seconds; larger sets use the governed replica in your lake.
Yes, through SQL: a saved pivot is a view with AS OF, and its display rules export with it.
It is the master-data platform's analytical layer. Planning systems keep their engines; they receive the golden dimensions from Lima and can read the pivots through the API.
Bring two systems that hold the same customers, suppliers or materials. In one working day we connect them, run the match on live rows, review the queue with your steward and publish the first golden view with its API.
