One of things I keep hearing over and over again is having ad group name and keyword correlation in order to increase your quality score. Although it isn’t a large factor in the long run (imo) it can definitely play an important role in setting your initial quality score. With that said, the easiest way I’ve found to find relevant ad group names is to simply find out the word frequency within your keyword list. For example on a job related keyword list, this was partially the results:

[jobs] => 1990
[job] => 863
[in] => 805
[resume] => 633
[salary] => 544
[employment] => 282
[for] => 257
[agencies] => 239
[resumes] => 190
[com] => 187
[monster] => 185
[canada] => 178
[online] => 169
[a] => 169
[uk] => 158
[london] => 154
[technician] => 145
[free] => 137
[of] => 132
[courses] => 125
[search] => 121
[salaries] => 117
[description] => 117

From this I get an instant idea of what ad group names I need to make and I’ll even take this list deeper and pull out the word frequency for only the keywords containing “jobs” to find subkeywords to build even more tight knit ad group names.

Here is a little snippet of php code that will produce the same results with your keyword list (stolen from: http://snipplr.com/view/2225/php-tag-cloud-based-on-word-frequency/ and modified slightly) :

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24

<?php
 
// Store frequency of words in an array
 
$freqData = array();
 
$keywords = file_get_contents('yourkeywordlist.txt');
 
// Get individual words and build a frequency table
 
foreach( str_word_count( $keywords, 1 ) as $word )
 
{
 
// For each word found in the frequency table, increment its value by one
$word = strtolower($word);
array_key_exists( $word, $freqData ) ? $freqData[ $word ]++ : $freqData[ $word ] = 0;
 
}
 
arsort($freqData);
 
print_r($freqData);
?>