Google Trends Reveals Nearly All

June 26, 2008
by Ian McIntosh

How useful would it be to know how much search volume each keyword received? You would be able to plan entire search campaigns around this information, plan traffic growth, increased revenue and ultimately the overall return on investment. Fantastic, Google have now opened the doors, allowing us to do just that.

Recent Development

Google have recently made an addition to their Google Trends feature, by attributing a scale number to each keyword. If you compare two keywords, one number will be 1.0, and the other will be either above or below 1.0 depending on search volume. The new feature can be used to accurately compare search volumes Year on Year, Month on Month or an average since Google began collating historic data, often going back to over 4 years.

For example, imagine we are interested in the 2 keywords “Car Insurance” and “Home Insurance”. Using data from 2008, we see the graph below. This essentially means that on average, if the site ranking number 1 for Car Insurance is getting 1000 visitors per day, the site ranking number one for the term Home Insurance would only be getting around 150 visitors per day, assuming that the two sites had an identical click through rate.

Since AOL accidentally released the search history of almost a million of their users in August 2006, getting an idea of the click through rate for the top 10 search results. By combining this information with the number of searches from Google Trends, you can get a fairly accurate estimate of how many additional visitors you are likely to get by improving your natural search ranking by just a few places.

How can we use this data?

Let’s take the example of an online pet food retailer that would target “fish food” as their main focus to their SEO strategy, and that is moderately successful, currently ranking 5th and receiving 100 visitors per month from that term. Let’s also pretend that the click-through rate tool mentioned earlier revealed that if this site was ranking first, we would be getting around 870 visits per month. Now, using Google Trends (see graph 2 below) we can see that if we diverted our resource onto targeting “dog food”, we could get 870 x 2.26 visitors per month ranking number one, i.e. 1966 visitors. We could calculate how much traffic we could receive in position 1 – 20 for the term “dog food” using this tool again, and make a far more informed choice regarding where our time could be best spent.

We could also use this information alongside Google Trends For Websites to give us an estimate for the percentage of total traffic a particular keyword provides any given website. This could be useful for competitor analysis, to get an idea of how much non-brand traffic a competing site receives for example.

As was mentioned earlier, the trends graphs represent search volume which does not accurately or consistently correlate with click through rates. Click through rates can vary significantly between keywords for many reasons. Simple things such as title / description tags, competing sites, brand names, and the number of paid ads around the SERPS could all influence click through rates. Therefore, it is imperative to keep the users intent in mind when comparing keywords.

Accuracy of Data?

The data given on the trends graph has been compared with a large amount of search data stored by Latitude, and has proven to be very accurate, to within 10%. The impact of this addition to Google Trends is huge and widespread, as it allows search marketers to forecast search volume by keyword. Clearly, large online marketing agencies holding huge quantities of data are at a distinct advantage as they will be able to provide the most accurate traffic .predictions across almost any vertical. However, smaller companies and marketers can use this information to enhance their own keyword research. As the Google Trends product is still being developed, there will hopefully be more accurate data available in the future, and possibly more useful features added such as this one.