When we left off at the end of Part I, I was talking about how Google “normalizes” the data they show on the trends graphs.

Let me talk for a moment about data normalization and why Google would want to do that and why you would want to see the data normalized.

Let’s say New York City produces 50 billion queries to Google every day. And let’s say Fulton Missouri produces 10,000 queries to Google every day. With such a huge number of queries coming from New York, the little town in Missouri would never generate enough queries about anything to be statistically significant. They would never show up on the bar graph about any query, you would simply see the largest cities jockeying for position in the cities column. The same would hold true for the countries column.

Normalizing this data allows you to see each city’s data compared to itself. In other words, you see the percentage of queries for a particular topic for a particular city or country – that way you will see graphed the regions that have the highest interest, rather than the highest number of queries.

Since we are talking about regions, I want to point out that there is a regions drop down box below the graph, so that once you have actually issued your query, you can adjust it to see the results from only a particular country. This will further highlight interest in your query within a particular country. What I found interesting was that when I changed the region to “Italy” for my query that I showed in part I, I found that there were two languages, English and Italian represented in the Italian results. All this means is that their google results were set to English, but it was an interesting piece of information.

With that explanation out of the way, let’s get to the fun stuff…

In Part I I showed you how you can compare up to five terms separated by commas, but there is more, much more!

I’m going to start by diving right in to show you the power of this tool and why I’m so excited about it. Then we’ll go back to the beginning and I’ll show you step by step why this works and how you can get the same results.

It is possible to compare collections of words. My friends with the B&B also have a motel on the property.

Let’s say we are targeting our neighbor state, Illinois, for a marketing campaign. What should we target, the B&B or the motel?

This query will show us the answer: b&b | (b and b) | (bed and breakfast), motel | motels

After the query is run, I set the region to United States, and when those results are returned I can tweak the results even further by setting the state to Illinois.

This is the result:

You can see from these results that folks from Chicago, our target city, split their queries just about 50/50 between motels and b&b.  You can also see that motels win out over b&b’s hard in the height of the summer months.

Now you can see the power in this trending information! 

Let me explain the syntax on the query so you can duplicate what I’ve just done. I use parenthesis ( ) around keywords that are made up of multiple words. The space character is a delimiter that will cause each word to be interpreted separately. I don’t want that, so I enclose each keyword that has a space within parenthesis.

The pipe sign (unix jargon) | is treated as an “or”. In the first part of my query I want to tally all the queries for all the different ways to say “b&b” into a single line on my graph. I do that this way:

b&b | (b and b) | (bed and breakfast)

What this is saying is that I want to see the results for people who are looking for ‘b&b’ OR ‘b and b’ OR ‘bed and breakfast’ – tally all this in one line of the graph for me, please!

Now I want to compare this with a graph that combines the different ways to say motel:

motel | motels

Finally, I compare these two queries using a comma (which was what we learned to do in part one)

b&b | (b and b) | (bed and breakfast), motel | motels

Just in case this doesn’t quite have your wheels spinning yet, think about using this in conjunction with the ~ function of a google search – this search will show you what google thinks are synonyms for you search term.

If you are not familiar with this, type in ~bicycle in the google search window and look at all the terms in bold on the results page.

Now, if you are a ThemeZoom user, you have probably already jumped to the idea of comparing your different VMADs by using the synonyms selected for each theme to examine the trending and query volumes by region and over all.

In the next installment of this series, I’ll get into more on how to use trending with ThemeZoom, the problems with “normalization”, and even more Google trending options.

Sue Bell
www.themezoom.com