Think of flat files as being similar to the files in a file cabinet drawer---a collection of single records each containing standalone data. Relational databases are a collection of tables linked together using a common piece of data, such as an account number, and can be arranged to highlight specific information for ad hoc queries. A relational database is a scalable and query friendly tool that provides the ability to capture a wide variety of data types.
Advanced Data Structuring
A flat file is a text file that stores a single record per line with each piece of information separated by a comma --- its data structure is self-contained and limited. Relational databases, however, store data in tables that are structured for the needs of the data. Each table is made up of rows and columns, and each column can be designed to hold or restrict a specific type of data. Many tables can be created within a single database. The advanced data structuring capability of the relational database allows programmers and database builders to create more complex relationships between data.
For example, a flat file may hold a persons name, address, phone number and account number. A relational database can hold the same data in a table and hold transaction information for that account number in another table and payment information in a third table --- all of which can be linked together to create a bigger picture. The flat file structure does not allow the linking of information from flat file to flat file and can only be printed as it is.
Ad Hoc Queries
The data structure of the relational database is ideal for creating ad hoc queries. Once tables have established links between them, a user or programmer can retrieve related data as needed. Relational databases can assist business owners, managers and supervisors with quick query requests by collecting and displaying sales data, employee performance data or production data when needed.
On the other hand, a flat file must contain all the necessary information in a single file in order to respond with the data. Flat files would need to be designed to respond to queries, which means queries and flat files would need to be pre-designed. This process slows down decision making and useful inquiry into business processes and ultimately can hamper business growth.
Relational databases are scalable, meaning they can grow larger or smaller as needed and be accessed by more users when necessary. More tables can be added, more records can be placed in existing tables, and although a single record in a table is accessed by one person at a time, many users can access the same table and work within it simultaneously.
Flat files are not scalable. Placing a million records in a flat file will slow down the opening and closing of the file, plus flat files can only be accessed by a single user at a time, which slows down work processes. Flat files are not a good data storage choice in a fast-paced and growing business environment.