Hi, this is Scott Twitchell. I'm in Utah today on a train, I thought I'd take a moment to record a bit more on my conversations. In many parts of the world, there has been an emphasis over a number of years to build data warehouses. One thing about the difference between a financial system and a data warehouse is often data warehouses are said to be informational. As opposed to operational. There's a different computing pattern that happens in a reporting process than happens in an operational systems.

Often they have different timeframes involved for producing the outputs, and sometimes they have less criticality for producing those outputs. As we noted in a number of the prior episodes, though, the financial system has both operational characteristics and information and characteristics. It has a cycle that has to be completed very consistently. At times when people are building data warehouses, they view the financial system as something that is an operational system. The outputs from the financial system are something that are put into the data warehouse to help produce reports when they're needed. Much of the data that also feeds into the financial systems also is used to feed the data warehouse, often with just a different level of granularity, because of the financial systems operational nature, because it has to run consistently and perform these time sensitive processes.

The data in the financial system has often been aggregated in order to allow it to run faster to meet the timeframes involved to reduce the compute capacity needed. The data warehouse often has more detail and but it has less of that time sensitivity to it. Often the difference between these different feeds at the detail level for the data warehouse and the summary level on the financial systems are what creates often reconciliation issues. The reconciliation where the summary numbers contained in both of these environments do not match, raising questions about which ones are accurate and which ones are not. Another key aspect of the difference between the financial system and the data warehouse is a double entry accounting system. process that often feeds a data warehouse is an ETL process.

This is often seen as a single record in a single record out process a single record from the source system becomes a single record in the data warehouse. In contrast, the financial system has a double entry system, a single record a single transaction and the source system becomes a debit credit later on in the financial system. So a financial system is fed by something that could be called the accounting rules engine Rather than an extract transformation logic, accounting rules engine turns a single transaction into a debit and credit to transactions into the financial systems. These differences create differences in how the data warehouse and the financial system represent their data. These differences cause questions as to where should when which information is correct, which summaries are the proper measurements, kind of approach to solving this we'll talk about in later episodes when we talk about lowering the level of detail in the financial system.