Cicero breaks down the process of retrieving and digesting the information into several smaller groups that closely align with packages purchased from Bloomberg. For example all of the equity files are contained in a single batch group. For each group of files the following methodology is used to load the local database.

1. Preparation - Cicero verifies that all of the necessary database connections and local directories are in place before any work is done. Additionally the fields.csv may be downloaded to insure up-to-date field information.
   
   
2. FTP - Cicero connects to the Bloomberg FTP server using your username and password and downloads the files in the selected group. Bloomberg provides multiple FTP servers for your use and Cicero will fail over to an alternate server if the primary server is not available.
   
   
3. Decryption/Unzip - Cicero methodically decrypts each of the files received from Bloomberg and upon completion of that process unzips each of the files to a directory.
   
   
4. Transformation - Cicero analyzes the unzipped files and prepares them for loading into the database. Examples of transformative steps that Cicero performs are translation from 23.7MLN to 23700000 and looking up any missing column headers.
   
   
5. Database Load - The database load step of the Cicero solution can be broken down into multiple phases:
   

1. Create database table (if necessary). If the needed database table does not exist on the host server, Cicero automatically creates one.
   
   
2. Add columns to the table. If the new data from Bloomberg has an additional column, Cicero will automatically create that column and post fill it with nulls. If the new data from Bloomberg is missing a column then nulls are added for that column value to preserve any old columns.
   
   
3. Load the data into the table.

Additionally the Cicero product offers the following tools: