Focus on what matters – save time with our new Data Anomaly Detector for Finance
Working in the financial department is hectic in general, let alone during the monthly or quarterly closing of books. Much of the time is spent on manually ensuring that the integrity of financial reporting meets the company and regulatory standards.
Profit Software’s Data Anomaly Detector for Finance helps any company to save hours of manual work by automatically analyzing the integrity of financial data collected from sources like general ledger, accounts payables or receivables. Not to mention the value of correctness in EOY reporting.
Our Data Anomaly Detector is using state-of-the-art statistical methods for detecting anomalies and level changes in financial data. Analysis can be done for combination of general ledger’s dimensions such as cost centers, accounts, products and customers.
Data Anomaly Detector for Finance is a sophisticated model based on statistical analysis. It detects anomalies and changes in data automatically. The results can be easily visualized using any modern BI tool.
The solution is easy to deploy as per your preference: Anomaly Detector is available on Microsoft Azure cloud or it can be deployed on-premise.
In order for Anomaly Detector to work the solution requires either Azure Machine Learning or Microsoft SQL Server (2016 or newer) to be installed.
Let us show you how it works
We understand that we must prove the value of the solution. Therefore we are proposing to proof-of-value phase consisting of the following steps:
- Kick-off meeting to determine your goals
- Extract the data from your ERP system
- Environment configuration to setup chosen dimensions
- Load data to Data Anomaly Detector
- Visualize and analyse the results
- Findings and next steps
Basic Requirements for the PoV
Solution can be run in cloud (SQL server 2016 in VM) or on-site (SQL Server 2016 and later). R Services and Analysis services are required.
Data extract as a CSV file delivered by customer
- Containing a minimum three years of data
- Containing dimensions such as month, cost center, account and vendor
- One measurement (actual)
- all dimension combinations require actual figures (at least zero value)
- UTF-8 encoded, semicolon as field separator, comma as decimal separator
Contact firstname.lastname@example.org to find out how your accounting team can save hours of manual work every month.