If the specific type of contents is already known before starting data compression, a suitable procedure could be choosen. Any type of contents has an immanent structure offering several opportunaties for compression algorithms. In the following some examples are listed to explain how specific characteristics can be utilized in practical use:

Text: The frequency of letters in a common text is varying. A compression could be achieved if frequent letters will be encoded with a shorter code length.
Pictures: The parts of a picture differ in complexity. This could be very low especially in monochrome parts or regular patterns. The encoding of those parts require less efforts in contrast to parts with more detailled contents.
Audio: Audio data also vary with regards to its complexity. A single tone of a specific frequency only requires a simple coding structure. The complex sound of an entire orchestra needs a more detailled modell especially if all instruments are involved.
Video: The visual part of video data consists of consecutive pictures. In most cases only a small part of them will change from picture to picture whereas the rest remains as it was. A typical example is a news speaker sitting in a studio in front of a fixed background. Only changes regarding the speaker has to be encoded. For the remaining parts it is sufficient to encoded them once until the next cut will follow.

< ^ >