SEO

Using AdWords to Test Key Phrase Overall Traffic

I recently wrote a post over at SEOmoz talking about how to forecast key phrase traffic. And when I say traffic, I mean the pool of total searches for one specific key phrase. From this pool comes traffic for paid and natural traffic.

With all the changes in the Google Keyword Tools, we are all left with a sense of “what do I do now?” One of the recommendations is to run an AdWords campaign to test key phrase traffic. If you have a specific key phrase that you need to know actual traffic on over time, there are some specific steps to get those numbers using AdWords.

Step One: Get an Account, etc.

If you aren’t already advertising with Google, open an account and all that jazz. With a recent client we just started a new account and I used a promotional code for $100 off for new accounts. We’ve spent more than that, so no skin off Google’s back.

Step Two: Pick fewer than 25 Key Phrases

Why? You are going to want to look at impression share with this test, so you can only have one keyword per campaign. Impression share is at the campaign level only (GOOGLE!!! Can we change this?) and an integral part of the test … and you can only have 25 campaigns per account.

Step Three: Create Your Campaigns/Ad Groups/Ads

This should not take long. Put your chosen key phrase into an ad group as an exact match and write a good ad around the keyword. You still need a good/quality/relevant ad and landing page, but not too good … unless you are looking to use this as true advertising too. If your ad and landing page are not relevant your ad can be shut down for quality score.

Step Four: Set Bid and Position Preference

All you need to do is show on Page 1. Remember that the only thing we want is the number of impressions, not necessarily clicks, but those will occur if everything is written right.

Step Five: Let it Run and Catalog

Keep up with the traffic and take into account impression share. To find impression share, you need to change the columns on the campaign view. Find out more including specific instructions on how to do it here at AdWords Help. Enter information on impressions and lost impression share. Then calculate the number of impressions per day.

From here you can go a step further and check the ranking of your organic listing. Use that rank to determine about how much of the traffic your organic listing should get. Then compare to your analytics to see if things are lining up right. If the traffic isn’t lining up, you might look into your title and description on the page to increase organic CTR.

Here is a sample spreadsheet to track things. Let the campaign run for a month if you have the time and money; otherwise, give it at least two weeks for a good look at traffic patterns.

Keep an eye on your quality score and percent of traffic missed using impression share. The higher percent of traffic you miss, the more something is off. Check out your quality score and adjust your ads. If that is okay, adjust you bid to ensure that your ad is showing well.

Once the test is done, compare with the keyword tools, and I think you will be surprised. Ours were off by half on two key phrases. We saw half the impressions that Google reported … in the NEW tool … on exact match. Remember, always take numbers with a grain of salt when forecasting. Grab as much data as you can and make sound judgments. If the numbers sound too high, or too good to be true, they probably are.

9e03625ddbbe84607a06e47104058deb 64 Using AdWords to Test Key Phrase Overall Traffic
Kate Morris is an SEO Consultant for Distilled Consulting in Seattle, WA. You can find her on twitter @katemorris.
9e03625ddbbe84607a06e47104058deb 64 Using AdWords to Test Key Phrase Overall Traffic

Latest posts by Kate Morris (see all)

You Might Also Like

Comments are closed.

13 thoughts on “Using AdWords to Test Key Phrase Overall Traffic

  1. This is just an analysis. May be, this will work but not sure. The mantra is simple, keep testing in as many ways as possible. After all, marketing is all about trial & error, experiments and risks….:)

  2. I found your post informative and also follow your suggestion and tips.Well google adword tool is used for searching keywords but now it test the keyword performance and traffic its good to heard that one.

  3. Your post and attached spreadsheet assumes that the impressions garnered from a paid search campaign can be wholly transferred to natural search. What about the traffic 'lost' to paid search results? 100 impressions seen on a given paid kw cannot be distributed out amongst the top 20 natural search results. You would have to first factor the 100 down by the natural/paid split, say 70/30, then distribute the 70 impressions amongst the natural results.

    Also, a flaw with the Chitika research is that it doesn't differentiate brand v. non-brand. Of course, branded searches will result in top positions, making the first results top-heavy in terms of CTR. The distribution of clicks will surely look different when branded searches are removed from the mix.

    My 0.02 worth…

  4. i still like to use few paid softwares for good niches. but i liked the new tool by gogole which is Google Keyword Planner. its new with better techniques. well ! after all its Google. but you can not rely only on adword keyword tool! any other suggestions would be greatly appreciated ! and thanks for this information :)