Marketing Channels Processing Rules - Internal Traffic

[jay_patel]

For some time now I’ve been getting more than expected internal traffic in my last touch marketing channels reports. On an average week we get about 9%-10% of our traffic from ‘internal’. We do have 5 international sites (language/country specific), all of which are linked together via a footer link, and are included in our internal URL filters. But I have a hard time believing that we getting a tenth of our visitors are moving between sites.  I believe I have followed best practices (screen grab below) on setting up my processing rules.

Anyone see an issue with my processing rules or any ideas on why I would see such a large share of my categorized as internal?

 [img]processing rules.png[/img]

[brian_au · Accepted solution]

Hi Jay,

The below Laura Chase Adobe Marketing blog post is a great guide to troubleshoot Session Refresh (Internal Traffic):

http://blogs.adobe.com/digitalmarketing/analytics/marketing-channels-5-steps-to-troubleshoot-session...

To review since last-touch Session Refresh (Internal Traffic) with no overwrite can only occur if it was also the first-touch, the below scenarios explain how Session Refresh could be a first-touch channel. I would recommend looking into how/if the below is occurring to inform if the Processing Rule needs further refinement or to be broke out into additional sub-rules:
• Scenario 1: Session timeout
• Scenario 2: Not all site pages are tagged
• Scenario 3: Redirects
• Scenario 4: Cross-Domain Traffic
• Scenario 5: Long entry page load times
• Scenario 6: Adobe Analytics cookie deleted mid-visit

Best,

Brian

[heather_gassman]

Jay - We have had similar problems with our attribution and I believe it is because we have been attracting more first-time-ever visitors and the very first page you see on our site has higher page-load-times for all of the cookies and tags than for repeat visitors.  Thus, I believe that these users are actually navigating to the 2nd page of the visit before our last-to-load s_code tag has time to capture the user's first page.  

To Adobe Analytics, it seems that the 2nd page of the visit is actually the first page and the traffic source is tagged as "Internal Referrer".  I am planning to alter my Marketing Channel Rules for these cases to look at the query string parameters of the Referring URL (which is an internal URL) so that I can properly attribute the channels in all of the cases where we were using query string parameters to set it in the first place.  At the same time, I'm going to try to work with IT to decrease the load times if possible for the first-time visitors.

Hopefully that helps!

Heather G.

[jay_patel]

Thanks, Heather.

Interesting, never thought about this. I know our desktop experience is fairly quick according to the tests we have run, but we do lag in mobile. Have you implemented the changes to the processing rules yet? Curious to see if this helped reduce the internal referrers for you.

Also, what I found recently is that that traffic sent to us from Facebook (within the mobile app) seems to be sent as internal vs. direct. However, if we add tracking variables to all links posted on FB, it solves the issue.

[PeterZH12]

Hi Brian,

Could you help explaining how scenario 6 would happen? It would be very helpful.

Thanks,

[jay_patel]

Thanks, Brian.

[brian_au]

Hi Peter,

The Adobe Analytics cookie could be deleted via a Browser action:

http://www.aboutcookies.org/DEFAULT.ASPX?page=2

Best,

Brian

[PeterZH12]

Thank you Brian! In certain days we are seeing non-persistence cookie visitors are higher. Does that mean transactions/sales from these visitors are not being captured in Omniture? One example is May 19th: we saw a dip in sales in Omniture, but backend data warehouse that processing orders did not show the dip. Could this be caused by Omniture cookie deletion? Why would some days cookie deletion is significantly higher than other days?

Thanks,

[brian_au]

Hi Peter,

Non-persistent cookie visitors should not impact orders tracking as that is a page level event. Also all visitors regardless of cookie status will be counted as visits in v15 (see explanation below via Ben Gaines). I would recommend you investigate and review the traffic types and trends on the days when you have discrepancies across your two reporting systems.

Best,

Brian

http://blogs.adobe.com/digitalmarketing/analytics/15-for-15-improved-metrics-and-logic/

>>

1. Vis­its for Non-Cookied Visitors

Site­Cat­a­lyst 14: Site vis­i­tors who do not accept per­sis­tent cook­ies via their browser are not included in the Vis­its total in any report, at either the site-wide or line-item level. This also applies to pathing data; for exam­ple, users who block cook­ies are not included in Bounce Rate cal­cu­la­tions, the Fall­out report, etc. Note that Dis­cover has always included Vis­its and pathing data from non-cookied vis­i­tors. This occa­sion­ally led to a sit­u­a­tion where your Unique Vis­i­tors could be higher than your Visits.

Site­Cat­a­lyst 15: All vis­i­tors, regard­less of cookie accep­tance, are included in Vis­its counts and pathing data.

What does this mean for me? First, it means a more com­plete and accu­rate view of many met­rics. Of course, a visit is a visit, and, ide­ally, cookie accep­tance should not deter­mine whether a user’s time on your site counts as a visit. Sec­ond, it means an increase in the Vis­its met­ric in Site­Cat­a­lyst, although the degree of this increase varies depend­ing largely on the type of imple­men­ta­tion that you are using. A sam­ple of data from our beta cus­tomers show a mean increase of less than 0.5% for first-party cookie imple­men­ta­tions, and an increase of 5–12% for third-party cookie imple­men­ta­tions (due to dif­fer­ences in cookie accep­tance rates).

Keep in mind that this increase will also affect cal­cu­lated met­rics that use Vis­its or a pathing-related met­ric (such as Entries, Aver­age Time Spent on Page, etc.) For exam­ple, many users have defined “Con­ver­sion Rate” as Orders divided by Vis­its. After upgrad­ing to Site­Cat­a­lyst 15, you may see a minor increase in Vis­its with no cor­re­spond­ing increase in Orders, lead­ing to what some could view as a small decrease in Con­ver­sion Rate (which, as any ana­lyst will tell, can often raise red flags).

This also means that you should never see your Unique Vis­i­tors count exceed Visits.

How should I pre­pare for the change? You can deter­mine the increase in Vis­its (and pathing met­rics) that you are likely to see post-upgrade by log­ging in to Site­Cat­a­lyst 14, going to the Site Met­rics > Vis­i­tors > Daily Unique Vis­i­tors report, and click­ing on the “Per­sis­tent Cook­ies” fil­ter. This will break out non-cookied vis­i­tors from cook­ied vis­i­tors in your exist­ing data; you can divide [non-cookied vis­i­tors] by [total vis­i­tors] to deter­mine the per­cent­age increase that you are likely to see when you upgrade to Site­Cat­a­lyst 15. Then, we rec­om­mend work­ing with your inter­nal cus­tomers to help them under­stand and pre­pare for this improve­ment in this com­monly used metric.

[PeterZH12]

Thanks, Brian!

[jamesc82861453]

Hi Brian,

I have same issue. Suddenly my organic traffic has reduced and internal traffic has increased. I did go through the article. But my concern is, till previous month it was all good but suddenly starting October, it is reducing with no potential changes in Adobe Admin Rules or Tagging in last 12 months. Where can be the potential problem?