Google Insights: Measure your marketing strategies

25 09 2008

Nowadays when most of us tech-savvy people read about something in the papers or hear about it from a friend, the first thing we would most probably do is Google it. If you think about it, statistics of search keywords on Google is actually quite a good measure of your company or product’s popularity among netizens which is a huge proportion of the population in most developed countries. Consequently, it is also a good measure of how effective your new marketing campaign has been.

I don’t know when Google released this new feature but I’ve only come to know about it last week while trying to get hold of some Google search statistics. This new feature is Google Insights which allows you to analyze trends of Google search keywords with enough volume. It is so comprehensive that you can use it to narrow search statistics down to not only countries, but also states and cities. I was quite skeptical about the trends shown in the statistics at first but decided to test it out by checking the trending for “David Cook” in the past 12 months.

The statistics look very accurate with the trend peaking during the May-June 2008 period when he was competing in the finals and crowned American Idol.

Seeing how accurate the statistics can be, I decided to use it as a measure of how effective some marketing strategies of products I’m familiar with is. A good example would be the popularity of the HP TouchSmart PC at the period when HP had the Bloggers’ Pre-Launch event. The event happened on the 24th of June so I took a look at the trend from June 2008 – August 2008 for the search term “HP touchsmart” with “Singapore” as the location.

If you scroll over the peaks and troughs in the graph above (within Google Insights), you will see the following data:

Date Search Index Comments
Until 22 June 0 Before Blogger event
29 June 70 After Blogger event + Launch in Berlin – Rise observed
6 July 82 Further rise observed (more posts about blogger event?)
13 July 69 Fall despite bloggers hosting competition on their blogs
20 July 82 After official launch in Singapore. Rise observed
24 August 100 Rise observed all the way until 24 August

Between 29th June to 6th July there was a rise in the trends, showing that the blogger event may have do HP some good by getting more netizens to know about the HP TouchSmart. However, there was another wave of bloggers blogging when HP hosted their “Win a TouchSmart” Competition but this time, a fall was observed (13th July) based on the assumption that all bloggers who attended the Blogger Event also blogged about the competition. This sort of shows that the earlier rise from 29th June to 6th July was more because of the HP TouchSmart launch in Berlin than the bloggers from the Singapore blogger event disseminating information on their blogs. 20th July onwards, the TouchSmart has been receiving a steady rise in popularity showing that the “old media” marketing strategies have been effective in reaching out to many more netizens.

We can see from the above that the Blogger Event may not have been successful with other factors like the launch in Berlin and the “Old Media” channels having a greater outreach to netizens. However, just like all statistics, the above analysis were based on many assumptions like during the official launch period, HP did not employ other “New media” marketing strategies that may have complemented and boosted the rise in trending and that they were doing a lot of heavy old-school advertising. The point here is, instead of estimating the influence a blogger event has on a product launch from the number of blog postings, a blog’s authority rating on Technorati, etc etc, it’d be better to maintain all other marketing projects constant and watch how the trending is affected at the period the event is held.

I’m not saying that blogger events, social media nights and new media marketing are insignificant but it would be more efficient to measure which type of events(including activities done during the event) and strategies(like starting a competition on Friendster or creating an app on Facebook) worked for a product and which ones did not. Subsequently, marketing strategists who are aiming to use more digital means can also have a better feel of what is more effective for their company / product in reaching out to their specific target groups than blindly using strategies others have used. As the saying goes, not every shoe fits everyone. You have to cut the shoe to fit your own feet.


Note: There are many other statistical tools out there which you can use for various purposes like Facebook Lexicon to view popularity of keywords on Facebook walls. This post was written to highlight an example of how a tool like Google Insights can be used to measure a popularity of your product/company among netizens.




3 responses

25 09 2008
Melvin Yuan

Hi Su Yuen,

Thanks for this thoughtful analysis!

The event in Berlin on 10 June was a global one, so understandably no one (very few) would search for “HP TouchSmart” in Singapore immediately after that. And your Google Insights chart reflects that.

What your chart showed – is that from 22 June onwards – until 29 June, there was the initial spike in “HP TouchSmart” in Singapore searches. And this is because, on/around 22 June, we started revealing that the blogger event was going to be about the HP TouchSmart. The event was on 24 June, so my hypothesis is – people talked about the HP TouchSmart Blogger event and the conversations on the HP TouchSmart continued after.

The level was sustained and given a boost when, on 4 July, the blogger contest was launched. Hence the increase that you observed… leading up to 6 July.

Naturally also, there would be a slight dip to a peak (cycles) after the contest was launched. And it was revived (possibly with other forms of communications).

So – the data indicates a direct correlation between the blogger event and contests with spikes in local awareness of the TouchSmart. It wasn’t the global launch in Berlin that did it – or we would have had the spike start from 10 June (not 22 June). (Note that the the local newspapers published the stories from as early as 12 June onwards…I’m not sure of the exact date now, but I’m pretty sure the local papers covered it shortly after the event.

All that said… the main point is – nothing beats having friends like you and the rest of the gang joining in, and being part of our product launches! So thank you 🙂 and see you at the next one!


25 09 2008

@Melvin: Hey thanks for the alternative analysis! Very insightful!

After reading your analysis, a new thought came to my mind! I have a hypothesis about the increase starting from 22nd June instead of 10th June from the global launch due to more people around the world who heard about Berlin, alongside heavy advertising by HP in both Berlin and other parts of the world are starting to check out the product and blog about it, and slowly it reached Singapore readers. Also, the international gadget/tech magazines may be taking some time to start putting the articles about the launch into their Asia region issues, which may again explain the lag.

Then again, I’m hypothesizing out of a lot of speculations and my knowledge of effects of various forms of advertising on internet search keywords is very weak. This is the first time I’m actually looking at actual search keyword statistics.

18 10 2008
Chen Chow

Su Yuen,

Fantastic analysis!!! 🙂

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: