Understanding Google Insights: You Can’t Estimate Traffic with It

SMS Text

Google Insights is an awesome tool for tracking and comparing trends. Most of us use it on a regular basis.

However the common misconception of the tool is that it’s a regular keyword research and traffic estimation utility. Well, to put it simply, this is not true:

1. The graph data DOES NOT equal search volume

Google Insights show “the likelihood” of people in each region to search for the given phrase:

This means that the number on a graph (roughly) = search volume for this word for each particular region / the overall searching population in this region and normalized to 100-scale graph (to be able to compare it on a graph)

Why is the data normalized?

Without normalization regions with higher population would always be on top simply because there are more people (and hence more searchers) there. This means we would never be able to find out that in some particular (smaller) region people like apples more than in other region with higher population.


If you search for “hairdo”, you might see Singapore ranked first and United States ranked 5. This does NOT mean more people search for “hairdo” in Singapore than in the United States. It just means the regional interest for hairdo is higher there. Conclusion: there’s no way to tell the absolute popularity of the term in any region using Google Insights.

Google Insights: Normalization

2. Downward Trend Does NOT Necessarily Mean Declining Search Traffic

Going back to the formula we discussed in the previous section:

Number on a graph = number of searches for the keyword in the region / number of all people who use Google search in this region

Now imagine for a minute that much more people start using Google search in this region, while the interest to the keyword remains the same (for example, people who start using Google search in this region belong to the group that is not interested in this keyword)? This would result in the downward trend on the graph but in reality the search traffic remains the same.

An Example?

…suppose the city of Melbourne, Australia, has 1000 internet users and 500 of them (or 50% of internet users in Melbourne) searched for the term spring festival in October. In November, 500 more internet users moved to Melbourne, but none of them knew about the spring festival, so no one out of that group searched for that term.

Here’s a forum thread that discusses the downward trend for many generic phrases like “loans”, “poker”, “insurance”, etc. Here’s how this is explained:

…a decline on the graph means a declining SHARE of all possible searches. To show a flat graph, a given keyword would actually need to be increasing in volume — assuming the total volume of all searches is increasing, and for most areas of the world this has been true.

Google Insights: Downword trend

Most Useful Articles from Help Section:


Ann Smarty
Ann Smarty is the blogger and community manager at Internet Marketing Ninjas. Ann's expertise in blogging and tools serve as a base for her writing,... Read Full Bio
Ann Smarty
Get the latest news from Search Engine Journal!
We value your privacy! See our policy here.
  • And if i compare two keywords of a particular region and I’ve the approssimative traffic data for one of those keywords?
    For example I know that the keyword “Piatti” in Italy brings 200 visits/day to the first search result. Then I set “Piatti” in Insight with not “All the World”, but with “Italy” ad region and set another keyword, for example “Lavelli”.
    With a comparison I see that (“Piatti” Search Volume)/(Lavelli Search Volume)=9.4.
    Can I estimate that Levelli brings to the first search result 200/9.4=21 visits/day?

    Paolo Dello Vicario

    PS: Sorry for my bad english

  • Ann:
    Great explanation! I use insights a lot, especially for regional clients as raw, worldwide numbers from tools like WordTracker can be deceiving. You’ve elegantly pointed out that Insights can be deceiving too, unless you carefully consider what the “normalized” data is really telling you.

  • Hmm.. very insightful indeed. It really sheds light on some VERY common misconception. Great post.

  • Hmm.. very insightful indeed. It really sheds light on some VERY common misconception. Great post.