Does Google Instant ACTUALLY Affect the Long Tail? (Free Downloadable Spreadsheet!)

Wednesday, September 15th, 2010

So last Wednesday we were all introduced to Google Instant, a new way to browse the SERPs that will dynamically generate results as you type based on what Google thinks you are searching for.  Aside from people crying out that SEO is dead (it’s not), the most common feedback regarding the new Google functionality seems to be that it will severely limit the benefit of ranking for long tail queries.  But is there any merit to these concerns?  Will long tail searches actually be curtailed now that Google can read our minds?  That’s what I wanted to find out!

The insights shared by some of the greatest minds in the search industry have been very enlightening, but these are still human insights.  I wanted to turn to cold hard data, which is the consummate authority when it comes to analyzing the impact of different Google updates.  To steal a line from the Dark Knight’s Harvey Dent, I wanted to defer to that which is “unbiased, unprejudiced, fair.”

In order to do that, I put together a process that will identify what percentage of the total keywords driving traffic to your site are 1-word phrases, 2-word phrases, and so on and so forth up to 7+ word phrases.  By comparing this data before and after the Google Instant launch, you’re able to see if your long tail traffic (or short tail, for that matter) was positively or negatively altered as a result of the change in SERPs. This strategy let’s you take an “Analytics First” approach to determining the impact, and is available absolutely free for all our valued readers here!

This spreadsheet is broken down into 3 tabs, “Keywords Before Instant,” “Keywords After Instant,” and “Comparison.”  In order to conduct your own analysis, all you will need to do would be to export your referring keywords from equal time periods before and after the Google Instant launch (9.8.10, for those keeping score) and add them to the appropriate tab under “Referring Keywords.”  Be sure to only export Google as the Instant update will obviously not have affected other search engines.

**NOTE** A helpful tip for exporting keywords: If you append &limit=5000 to the URL string in Google Analytics it will let you export 5000 keywords at a time as opposed to the traditional maximum of 500.  There will be no visible change to the Google Analytics interface but your CSV export will reflect the increase.

Once you’ve added the keywords to the appropriate tabs, the Excel gods will take care of the rest!  You can then see how many of your referring keywords were 1-word phrases, 2-words, and so on as well as what percentage of your total keywords were 1-word, 2-words, etc. before and after the update.

Finally, a visual graph is generated that gives you a side-by-side look at how the composition of your referring keywords changed (or, perhaps, didn’t change at all) as a result of the update.

I would like to point out a few points to keep in mind as you dig through the data:

• Word count is calculated according to the number of spaces included in the phrase.  That being said, please keep in mind that the following examples will be recorded as 1-word phrases:
• URLs – Ex. www.seerinteractive.com
• Phrases with dashes – Ex. Back-to-back
• Underscores – Ex. Brett_Snyder
• You should take special care to compare equal time periods.  This includes comparing equal number of days (7 days before the launch compared to 7 days after) as well as equivalent time periods (Mon-Fri before the launch compared to Mon-Fri after).  This ensures that other variables such as seasonality are neutralized.
• Google Instant is not yet available to everyone,  so some users may still be searching “the old-fashioned way.”
• Google Instant is not enabled for users searching within a toolbar.

Finally, I would like to reiterate that this spreadsheet was designed to help the search community use data and analytics to determine the impact of Google Instant on long tail vs. short tail traffic.  We’re very interested to hear how your traffic numbers compare and I will be following up with some of SEER’s insights in the next couple of weeks after more data has been aggregated.  I encourage you to post your results in the comments so we can pool our analysis and come up with some veritable proof to put the question of Google Instant’s impact on the long tail to bed for good!

Once again, the spreadsheet is available for download absolutely free by clicking on this link.  I look forward to any questions, comments, or concerns!

Posted in SEO | 22 Comments »

Important Exception to Google Analytics Last Click Attribution!

Wednesday, September 15th, 2010

Cutting right to the chase, Google Analytics uses last click attribution with ONE important exception, direct traffic.

*Think you already know everything there is to know about Direct Traffic and Last Click Attribution? Skip down to “How Does This Affect Me?” and you might be surprised!

So why does Google Analytics handle direct traffic differently? What can this mean for your data? Understanding exactly what direct traffic is and how Google Analytics tracks it can not only help you to better understand your reports, it can save you a lot of time and confusion.

Direct Traffic

First let’s take a look at exactly what direct traffic is. Direct traffic is any traffic to your website that does not have referring information. Common examples of this include bookmarks, typing a URL directly into your browser, links in common twitter applications, instant message links etc. The most important thing here is to understand that direct traffic doesn’t mean the visitor somehow “directly” came to the site, but rather that Google Analytics has no information about where this visitor came from.

Last Click Attribution

Now let’s take a quick look at how last click attribution works. If I do a search in Google and click on an organic link, Google Analytics will record my referring information as Google Organic. Then if later that day I’m surfing the web and I see a link to the site I visited earlier and click on it Google Analytics will overwrite the previous referring information of Google Organic with Referred Traffic. In summation, the “Last” thing I click on will be the source that gets credit for the visit to my website.

Why is Direct Traffic Handled Differently?

Now that we know exactly what direct traffic is, and how last click attribution works, it becomes fairly easy to understand why Direct Traffic is the exception to the rule. The easiest way to demonstrate this is with an example:

A visitor does a search on Google and clicks on an Organic link, setting their referring information to “Google Organic”, and then the next day the visitor comes back to your site via bookmark.

If direct traffic worked the same way as the rest of traffic in Google Analytics, you would over-write the previous referring information (Google Organic) with the “Last Click” (direct traffic). The problem is, overwrite it with what? We established that direct traffic means Google Analytics has no information about the visitor. Since we have no information if we followed the last click attribution method we would overwrite our previous referring data (Google Organic), with nothing!

Clearly there is no benefit of overwriting some information with no information, so instead Google Analytics simply leaves the previous referring information. As a result for the previous example above Google Analytics would show two visits generated from Google Organic, and NOT one Google Organic and one direct visit. This is reasonable for Google Analytics; if not for Google Organic the visitor wouldn’t have found the site, so giving credit for both of these visits to Google Organic makes perfect sense.

How Does This Affect Me?

Everyday Reports

If I’m looking at my Google Analytics and I see I have 100 visits yesterday from Google Organic, that doesn’t necessarily mean that 100 people searched on Google clicked a link and came to my site. It can also mean that 90 people searched on Google today and came to my site, and 10 people searched yesterday on Google and reached my site, but decided to come back VIA bookmark.

Save Time and Confusion

While this may help us to understand our reports better, how does this understanding save you time and confusion? Consider the following examples:

Example 1

Ever turned off a paid search Campaign and continued to see the campaign send you traffic? Direct traffic is the Culprit! If someone clicks on a Google Paid Search ad their source will be written as Google Paid Search (technically cpc in GA). If the visitor then continues to come to your site by a “Direct” method, since the previous data isn’t overwritten they’ll continue to show up as a Paid Search visitor. Ever wasted time trying to figure out how your discontinued paid search campaign is still sending you traffic?

Example 2

You have a website which has just started a new paid search campaign. This website has a landing page specifically designed for this new campaign, and it is only reachable through Google Paid Search. So when you go into Google Analytics to see how this page is performing you are confused to see Yahoo Organic as a source / medium for this landing page. How is this possible?

This is where it gets a bit tricky!

First, someone does a search on Google and clicks on your paid search ad. Once on your PPC specific landing page the visitor decides that they’d like to come back to this page later so they bookmark it. The next day the same visitor is interested in something else on your site, but they happen to be on yahoo at the time. They decide to do a quick search for your brand name and click on a Yahoo Organic link to reach your main site. Later that week the visitor remembers the page they’d bookmarked and decide to go back to it via bookmark. When they follow that bookmark and land on your Paid Search landing page they will show up as a Yahoo Organic visit with your Google PPC page as their landing page!

Following the Google Analytics Referring information through this example:

1. Click on an Adwords Ad in Google. Referring information is now “Google Paid”.
2. Do a search in Yahoo and click on an organic link. Referring information is now “Yahoo Organic”.
3. Follow your bookmark to the Google PPC landing page. Leave referring information as “Yahoo Organic.”

Without understanding that this kind of behavior is possible simply due to how Google Analytics processes data, you might very well spend hours digging through your website and Adwords configuration trying to figure out how Yahoo Organic traffic was reaching your Google Paid Search landing page.

Have you ever seen direct traffic perform unexpectedly in your reports? Do you have any methods for handling or improving the information you do have about direct traffic?

We’d love to hear from you!

Posted in SEO | 2 Comments »

Wine Enthusiast partners with SEER

Tuesday, September 7th, 2010

Cheers! SEER has partnered with Wine Enthusiast, a premier seller of wine coolers and wine accessories, for an SEO project.

Posted in News | Comments Off