Create a grid on your graph paper that is 12 by 12 blocks.
Arrange the 12 pitches of the chromatic scale in any manner across the top row. Write one pitch per square in the grid so that there is a total of 12 pitches in the top row. Avoid using pitches that hint at major or minor tonalities. For instance, the first three notes shouldn't be A C E as those notes imply a minor triad.
Starting with the first square and moving diagonally down the row to the right, fill in the note for each square using the first pitch of the row created in step 2. If your note in step 2 was an A, then you will be left with a line of 12 A's diagonally from the top left to the bottom right of the grid.
Determine the transposition of the second row by taking the pitch in the second column of the first row and comparing it to the pitch of the second column in the second row. If there is a Bb in the second column of the first row and an A in the second column of the second row, then you would need to transpose all of the pitches in the second row a minor second down.
Continue transposing each row. If there is a C in the third column of the first row and an A in the third column of the third row, then every note will be transposed down a minor third from the notes in the first row.
Complete the transposition level of each row along the top and the left side of the matrix. For instance, if the first note is A, the transposition level is 9. The numbering of each row is useful for referring to a specific part of the row. For instance, to refer to the first left-to-right row in the matrix, you could refer to it as Prime 9 to indicate that A is the first note of the transposition and you are using the prime version of the row. There is a listing of other forms of the row in the Resource section of this article. Numbering of notes is completed by starting with C = 0 and moving up chromatically. For example, C = 0, C#/Db = 1, D = 2, D#/Eb = 3 and so on.
Copy the numbers from the top row to the bottom row and the left row to the right row of the matrix to complete the numbering system. We do this so that every row has a unique number and identifier, such as Inversion 0 or Retrograde Inversion 0. Following this process helps with the analysis of the piece. Similar to tonal music where an emphasis is on chords, in 12-tone serialism, numbers and forms of the row are used.
Sing or play a row created from the 12-tone matrix. You can choose any row that you wish, although most composers will start with the top-level prime version of the row. This helps to identify the initial base pitches for the entire piece.
Change any intervals that create patterns, sequences or tonal references. Since the goal of serialism is to avoid tonal references, it is crucial to omit these references. Avoid anything that implies tonality. By creating patterns, sequences and tonal references there is an implication of a tonal center. The implication of a tonal center is only advisable with tonal music. Serialism is essentially a form of atonality; the rules prevent the dependence on a tonal center.
Compose the piece using different forms of the row. It is possible to write using pitches that move left to right, right to left, top to bottom or bottom to top.