Automated Journal Loading & Validation

Three business units within a large ASX-listed financial services group needed to upload thousands of journal lines to Dynamics 365 each month-end. Using the native D365 Excel integration, each load took 8 to 12 hours - long enough that it had to be left running overnight on a staff member's laptop. When a load failed, month-end close was delayed, and the error feedback provided was limited enough that diagnosing the problem added further time.

I designed and built an Azure-based journal loading pipeline to replace this process entirely. Accountants simply dropped their journal file - Excel or CSV - into a SharePoint folder. The pipeline ingested the file into SQL Server, ran it through a Python-based validation process that checked balancing, field completeness, and master data alignment, then loaded the validated journal into D365 via the Data Management framework. Throughout, the submitter was kept informed via Teams, receiving detailed, actionable feedback if any issues were found.

The result was a reduction in load time from 8-12 hours to under 5 minutes - and the elimination of a recurring close risk. The validation engine was subsequently extended to cover automated journals from travel and expense systems, loan management platforms, and other group processes, giving a consistent, auditable control gate across all journal sources into the ERP.