Building a Single Source of Truth for Sales, Stock and Finance
When sales, the warehouse and finance each bring a different number to the meeting, the cause is usually structural: nobody agreed where each figure lives or who owns it.

Sales says the month closed at 410,000 dinars. Finance says 386,000. The warehouse says it shipped more than both. Everyone is partly right, and the first half hour of the management meeting goes on reconciling figures instead of deciding anything. A single source of truth is the fix, but it is often misunderstood as “put everything in one system”, which is neither necessary nor sufficient.
What a single source of truth actually means
A single source of truth is an agreement, backed by systems, that each important figure has exactly one authoritative home and one definition. Sales orders live in one place. Stock on hand lives in one place. Posted revenue lives in one place. Every report that shows those figures reads them from that home, rather than from a copy someone exported last week.
You can have several systems and still have a single source of truth, as long as each fact is owned by one of them and the others consume it. You can also have one ERP (enterprise resource planning) system and no single source of truth at all, if people keep parallel spreadsheets because they do not trust what the ERP shows.
Why the numbers disagree in the first place
When figures disagree, the cause is almost always one of four things, and it helps to name which one before reaching for a technical fix.
- Different definitions. Sales counts an order when it is confirmed. Finance counts revenue when it is invoiced. The warehouse counts when goods leave the gate. All three are valid. They just measure different events.
- Different timing. A report pulled on the 1st of the month and another pulled on the 3rd will differ because late invoices, credit notes and returns were posted in between.
- Copies that drift. Someone exports the customer list to Excel, adds a column for region, and that file becomes the basis for every sales report. Meanwhile new customers are added in the ERP and never reach the spreadsheet.
- Manual re-entry. An e-commerce order is typed into the ERP by hand. Every re-keying is a chance for a wrong quantity, price or customer code.
Only the last two are technical problems. The first two are decisions that someone senior needs to make and write down.
An illustrative case: a manufacturer in Sahab
Consider a mid-sized plastics manufacturer in Sahab that sells to distributors in Jordan and Iraq. It runs an ERP for finance and inventory, a separate spreadsheet for sales orders because the sales team found the ERP order screen slow, and a production planning workbook maintained by the factory manager.
Stock figures disagree weekly. The production workbook shows finished goods that have been produced but not yet booked into the ERP. The sales spreadsheet shows orders that were promised but never entered. Finance sees only what reached the ERP. Nobody is careless. The structure guarantees disagreement.
The first step in a case like this is a decision rather than a purchase: the ERP becomes the home for orders, stock and invoices. Then the ERP order screen is made workable for the sales team, production output is booked into the ERP at the end of each shift, and the side spreadsheets are retired once the ERP holds what they held.
Building it, step by step
- List the figures that matter. Usually fewer than twenty: orders, shipments, invoices, collections, stock on hand, stock value, gross margin, open purchase orders and a handful more.
- Assign each one a system of record and a business owner. The owner is a person who decides the definition and answers questions about it, not a systems administrator.
- Write the definitions. “Net sales equals invoiced value less credit notes, excluding sales tax, by invoice date.” One line each, agreed by sales and finance together.
- Remove the reasons for side copies. Ask why each shadow spreadsheet exists. Often it is a missing field, a slow screen or a report nobody built. Fix that, and the spreadsheet loses its purpose.
- Connect rather than re-key. Where two systems must share data, such as an online store and the ERP, use an integration or an automated import instead of manual entry.
- Point every report at the source. Management reports and dashboards should read from the system of record or from a reporting layer that copies it on a schedule, never from personal files.
Where a reporting layer helps
For organisations with more than one operational system, a reporting layer is often the practical middle ground. Data from the ERP, the online store and perhaps a customer relationship management (CRM) system is copied on a schedule into a reporting database, where the agreed definitions are applied once. Power BI or a similar tool then reads from that layer.
This keeps heavy reporting queries away from the live ERP and allows you to combine sources cleanly. The rule that keeps it honest is simple: the reporting layer never becomes a place where people edit data. Corrections go back to the system of record, and the next refresh picks them up.
Automated checks are worth adding at this stage. A nightly job that compares total invoiced value in the ERP with the total in the reporting layer, and alerts someone if they differ, catches broken integrations before they reach the board pack.
Making the truth stick
A single source of truth is part architecture and part habit. The architecture is the clear map of which system owns which fact and how the others receive it. The habit is the management team refusing to discuss numbers that did not come from the agreed source. Once that habit forms, the side spreadsheets fade on their own.
Our Enterprise Systems service works on exactly this: data architecture, system design, integration between platforms and dashboards built on clean data, so that sales, stock and finance finally bring the same number to the meeting.



