The cornerstone of most SEO/SEM strategies relies on robust keyword research and selection.

There is a large number of tools that help this process such as the free Google Keyword Planner Tool or other paid options such as SEMrush.

These are excellent for selection of preliminary keywords. They give a good sense of how much volume you could expect from the keywords and also a sense of the PPC competition through CPC reports (with higher CPC indicative of higher relative business values) .

But what if you wanted more?

Enter Google Prediction:

According to Google Where search predictions come from:

  1. Search terms you type. Relevant searches you’ve done in the past, if you’re signed in to your Google Account and have Web & App Activity turned on.
  2. What other people are searching for, including trending searches. Trending searches are popular topics in your area that change during the day and aren’t related to your search history.

and further Google elaborated:

Search predictions are generated by an algorithm automatically without human involvement. The algorithm is:


  1. Based on several factors, like how often others have searched for a term.
  2. Made to show the range of information on the web. You might get predictions related to many popular topics.

Hence if we do not login, it should be a good representation of what others are searching organically and also what might be of high relevance. Perfect for basing an SEO/SEM strategy around.

It is quite easy to use – just simply enter your keyword modified from a to z and very quickly you could see interesting suggestions. Like in the following example for vitamins:

We could target the issues that people looking for vitamins could solve and build useful content around where it overlaps with what our products can realistically help with.


Automating it Python is simple once you have Selenium setup, send the following requests if you are looking for “how to queries”:

alpha=" "+string.ascii_lowercase root="business how to " driver=startEngine("") inputElement = driver.find_element_by_name("q") for letter in alpha: inputElement.send_keys(root+letter) time.sleep(2) subdrivers=driver.find_elements_by_xpath("//div[contains(@class, 'sbx1')]") time.sleep(1) inputElement.clear() results=results+[i.text for i in subdrivers]

You could use run a gauntlet of queries ranging from commercial intent such as “where to buy…” in Singapore to find opportunities.

Do let me know if this worked out as good for you as it did for me in finding commercial intent!