by Daniela Araujo
Part I: Tracking Visitors
Careful analysis of online campaign performance data is an integral part of the PPC Account Manager’s daily routine. This includes tracking the effects of changes to the account, monitoring of long-term trends, and recognition of anomalies in data which warrant investigation. During this process, it is common to discover discrepancies in the data reported by Google Adwords and Google Analytics. Such discoveries can be unsettling to Account Managers, especially since both reporting systems are Google products. In this article, I will explain some possible causes for these discrepancies.
How Visitor Tracking Works in Adwords and Analytics
Adwords. Every click on an Adwords ad immediately triggers a request to one of Google’s servers before the user is redirected to the advertiser’s landing page. Each click is recorded by the Adwords system. Under normal circumstances, each click corresponds with a visit to the advertiser’s site (I will cover some exceptions later).
Primary Causes of Discrepancies in Visitor Tracking
2. Missing or Broken Tracking Code.
A common Analytics discrepancy can be caused by errors in the Analytics code placement. If code is not properly placed or is missing, Analytics won’t report visits – but Adwords will still report clicks. Analytics code is supposed to be placed in every page of the website, so if the webmaster doesn’t realize that a site has Analytics code across all the pages he/she could accidentally remove it. This is a very common mistake. Other potential misreads can be caused by extra spaces or missing pieces in the code. We strongly recommend cutting and pasting code into a notepad file if it needs to be sent to the webmaster so that these errors are avoided.
4. Incorrect Tagging of Links in Adwords.
Analytics uses a query string parameter “glcid” to associate a visitor with a campaign/adgroup/ad/keyword. So if the query string is incorrect or missing, discrepancies between Analytics and Adwords should be expected. Below are common situations that we have often witnessed:
a. Auto tagging is disabled. Google will tag URLs automatically if you enable the auto-tagging. This step should be taken right before campaigns are launched.
To enable auto-tagging:
1. Log in to your AdWords account
2. Click the My Account tab and select Account Preferences
3. In the Tracking section, click Edit
4. Select the Destination URL Auto-tagging check box
5. Click Save changes
b.Tagging was done manually and there are errors or missing pieces. Because some websites do not allow arbitrary URL parameters and serve an error page when auto-tagging is turned on, some advertisers decide to tag URLs manually. If this is the case, we recommend utilizing the URL builder tool to avoid errors:
c. URLs have redirects. For example, a PHP redirect in the landing page blocks Google Analytics code from launching and identifying the visits coming from Adwords. Google Analytics uses ‘gclid’ parameter in Destination URLs in order to identify visits from AdWords ads. The ‘gclid’ parameter shows up in the landing page URL when a user arrives at a website from an AdWords ad. For example, when you search for “xxx” and click in an Adwords ad, you see the url:
In case an advertiser is using redirects, you will search for “yyy,” then click on the Adwords ad and see:
http://example2.com/products_services/ (no gclid)
d. Bookmarked Links. Analytics uses the ‘gclid’ parameter in your Destination URLs in order to identify visits from AdWords ads. The ‘gclid’ parameter shows up in the landing page URL when a user arrives at a site from an AdWords ad. For example, if the site is www.xyxy.com, then when a user clicks on an ad, it will appear in the address bar as:
If visitors bookmarked the xyxy.com website along with the ‘gclid’ parameter, Analytics will record visits from the bookmarked
site as visits from AdWords ads. However, AdWords advertisers won’t be charged for these visits since it’s not an actual click
on the ad. Therefore, AdWords clicks will be lower than Analytics visits in the campaign under analysis.
5. Adwords reports Clicks. Analytics reports Visits. Clicks versus visits.
Clicks indicates how many times your advertisements were clicked by visitors, while visits indicates the number of unique sessions initiated by your visitors. When a user clicks on an Adwords ad and is taken to a landing page, the Adwords system records the click and Analytics records the visit. But if the same user navigates away from the landing page and clicks again on another ad reaching the same website, Adwords will report 2 clicks and Analytics will report only one visit or visitor.
6. 30-minute timer for session expiration.
If a user navigates away from a landing page and then returns within 30 minutes, Analytics will count it as the same visit while Adwords will count it as a new visit if the user came back through an Adword ad. On the other hand, if the user hits “back” 31 minutes after leaving the landing page, Analytics will record it as a returning visit (new visit, same visitor). A common comparison shopping behavior. For example: Maria is iphone shopping, sees an Adwords ad from iphoneplanet.com, clicks on it, but decides to research a little more before purchasing it. She navigates away from iphoneplanet.com and goes to cellphones.com. She discovers that iphoneplanet.com has a better price and after 29 minutes she clicks again on the iphoneplanet.com Adwords ad. In this case, Adwords will report 2 clicks while Analytics will report only one visit because she went back to iphoneplanet.com within 30 minutes. Analytics will consider that visit as part of the first session initiated 30 minutes ago.
7. Tracking code in landing page fails to execute.
Analytics code fails to execute when a user navigates away while the page is loading, clicks on the stop loading button or there is an error in the network causing Internet to fail. In these instances, visits won’t be reported.
Discrepancies and anomalies in web traffic across Google Adwords and Analytics should be expected, although such discoveries can be unsettling to Account Managers, especially since both reporting systems are Google products. I described common situations and potential reasons for discrepancies and listed common errors that should be avoided. Minor discrepancies are to be expected since these tools rely on different tracking methods. In Part II of this series, I will discuss discrepancies in transactions and revenue.
Category: Web Analytics