An-Nakel's approach is complete different from word-by-word translation.
The sentence is the basic unit for the translation process. An-Nakel is a rule-based processing system and operates on a knowledge database. Each sentence is analyzed deeply. An-Nakel determines the different syntagms and categories of words. It attempts to analyse words in context and then builds semantic links. The analysis process ends with an internal representation of the sentence. The target text is generated using the transfer method according to the grammar rules of the target language. In the case of translation from English to Arabic, the generated Arabic text can be displayed (printed) with or without vowels.

An-Nakel uses four on-line dictionaries:

The General dictionary contains words in common use with approximately 100,000 words and locutions. The Translation Memory database contains idioms, locutions and translated units. The Specific dictionary contains words used by specialists in a selected subject area. The User Dictionary contains words added or updated by the user.

The An-Nakel offers two operating modes:

- on line mode: the user may input and translate a document sentence by sentence and check the meaning of polysemous words;

- batch mode: the user submits the entire source document (file) and the target text is stored in a separate file.

In both mnodes, the target text can be checked or post-edited using the Alternative Meaning process.