Romanian companies have an advantage over Moldovan ones: part of the data is already structured, because electronic invoicing forces it to be.
That shifts the problem. It is no longer "where do we get the figures", it is "why does the accounting figure not match the sales figure".
Where the discrepancy actually comes from
Different moments. Sales report on the order, accounting on the invoice, treasury on the receipt. All three are right and none matches the others. Without a written definition, the meeting turns into an argument about numbers.
Parallel systems. The shop, the ERP, the accounting package, local Excel files in every department. Each has its own version of the truth.
Retroactive corrections. An invoice reversed the following month changes a figure that has already been reported. If the dashboard does not handle that, the history changes on its own and nobody trusts it.
VAT and classifications. The same sales look different with and without VAT, and comparing reports that use different conventions produces false conclusions.
What gets built
One set of definitions, written down. What a sale is, what margin is, which period closes a month. That document is worth more than the dashboard.
Automatic collection from every system. Including from accounting, where the data is already clean for compliance reasons.
Dashboards per role, on the same definitions. Sales, operations, finance — different questions, one basis.
Immutable history. The figure reported last month stays the figure reported last month, with corrections visible separately.
Threshold alerts. Margin falling on a category, a client whose orders are declining, stock that is not moving. Flagged when it happens, not in the quarterly report.
What we build
Automatic collection from every system, agreed written definitions, dashboards per role, immutable history and threshold alerts. Running on your infrastructure or on one you control.
Who it is for
Companies where the monthly report is assembled by hand and takes days. Companies where two systems give different answers to the same question. Companies deciding on instinct because the real figure arrives too late to be useful.
What is not worth building
A dashboard with forty indicators. It gets opened twice and abandoned.
The test: if you cannot write down the five questions it has to answer, it is not ready to be built.
Questions
Why do the figures not match between departments?
Because they measure different moments of the same process. It is not a system error, it is a missing definition. It gets solved in writing, before any chart.
Can you read from SAGA or SmartBill?
Yes. Each has its own interfaces, and the effort depends on which is the primary source.
How long does it take?
From a few days for one automated report to a few weeks for a dashboard unifying several systems with history kept.
Who owns the data?
You do. It runs on your infrastructure or one you control, and what that means is described here.