A while back, I created a useful script to help search for niche keywords on Google.com. I got the idea from AIR app that I found on the web but it kept crashing. It also had a bunch of needless features that only helped to slow down the process.
Let me explain what the script is doing so that you can understand how it helps to find niche keywords for you. I’m sure you noticed that when you start typing in text on the google.com website, it starts to give you a list of keywords that most relate to what you’ve already typed. The goal is to show you what other users search for often in order for you to find the most relevant data.
My script takes advantage of this functionality. By seeding the search with a word or two that relates to your niche market, the niche keyword tool adds 2 alpha-numeric characters after it and captures the keywords that Google returns. For example, if you seed the script with the keyword ‘how to’, my script will append ’00′ to the end of it and post ‘how to 00′ to Google and save anything that is returned. The script will continue this process until it’s exhausted all of the letter and number combinations (it’ll end with ‘how to zz’ in this example).
Before writing the keywords to a txt file, it’ll remove all duplicates and list them in alphabetical order. You’ll need to run the script multiple times with different keywords and pair them with Google’s Keyword Tool to find high-traffic, low-competition, low-cost keywords. Look for another post about how to get this information.
You’ll need to first install the Requests library from http://pypi.python.org/pypi/requests or you can modify the __getKeywords() method to use the archaic urllib and urllib2 libraries but they are seriously a pain in the ass to use!
Script usage example: python google_keyword_tool.py ‘my keyword’
and finally, the script:
import requests
import sys, os
from string import digits, ascii_lowercase
class KeywordFinder(object):
def __init__(self, keyword):
self.keyword = keyword
self.elements = {'hl':'en', 'q':self.keyword}
self.url = 'http://google.com/complete/search'
self.alphanum = digits + ascii_lowercase
def __getKeywords(self):
"""
gets the top 10 keywords for a given term
"""
headers = {}
headers['Accept'] = 'text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8'
headers['User-agent'] = 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10.6; rv:2.0.1) Gecko/20100101 Firefox/4.0.1 FirePHP/0.5'
response = requests.get(url=self.url,
headers=headers,
data=self.elements).read()
return self.__parseReturn(response)
def __parseReturn(self, response):
"""
parses out the cruft from the response and returns a list
"""
try:
response = response.split('window.google.ac.h')[1]
return eval(response)[1]
except:
print 'looks like Google is blocking your ip... <sigh>'
sys.exit()
def __status(self, char1, char2):
sys.stdout.write("Checking: %s %s%s\r" % (self.keyword, char1, char2))
sys.stdout.flush()
def __createFile(self, keyword_list):
filename = '%s/keywords-%s.txt' % (os.getcwd(), self.keyword.replace(' ','_'))
f = open(filename, 'w')
for k in keyword_list:
f.write('%s\n' % k)
f.close()
return filename
def recurseKeyword(self):
"""
recurse alpha-numeric characters following the keyword
"""
keyword_list = []
kws = self.__getKeywords()
for k in kws:
keyword_list.append(k[0])
for char1 in self.alphanum:
for char2 in self.alphanum:
self.__status(char1, char2)
self.elements['q'] = '%s %s%s' % (self.keyword, char1, char2)
kws = self.__getKeywords()
for k in kws:
keyword_list.append(k[0])
new_keyword_list = sorted(set(keyword_list))
filename = self.__createFile(new_keyword_list)
print 'Created %s' % (filename)
return new_keyword_list
if __name__ == "__main__":
if not len(sys.argv) == 2:
sys.exit()
keyword = sys.argv[1]
kf = KeywordFinder(keyword)
keyword_list = kf.recurseKeyword()
Enjoy!
