Referral Sites/URLs

Avatar
Gina Price
  • updated
Hello,

I was wondering if there are any help documents that get deeper into the understanding of the referrals. I am confused about my results.

I ran a report on our website (Landing Site A) and can see the referrals. But, there was a red flag for us there because a site (referral site) that should have extremely high referrals to Landing Site A was super low.

I then ran a report with specific include filter to try to grab URL information from the referral site. This report showed more referrals to Landing Site A. I am wondering why this would happen. Which report is accurate?

You can see that our referral site isn't even in the top 22.

Landing Site A Top Referring Sites example data
1 http://www.google.com 102,718
2 No Referrer 47,607
3 http://www.extension.purdue.edu 16,488
4 http://www.google.co.in 6,114
5 http://www.google.ca 5,884
6 http://www.bing.com 5,427
7 http://www.google.co.uk 4,251
8 http://www.google.com.ph 3,768
9 http://www.google.com.au 2,688
10 http://www.google.com.my 1,387
11 http://r.search.yahoo.com 1,346
12 http://www.google.co.za 1,338
13 http://www.ces.purdue.edu 1,258
14 http://www.google.com.mx 952
15 http://www.google.com.pk 936
16 http://www.google.co.th 675
17 http://www.google.co.nz 636
18 http://www.google.com.ng 576
19 http://www.google.ie 564
20 http://www.google.com.eg 530
21 http://www.google.es 470
22 http://www.google.com.vn 463

Now using include referrer filters
1 http://mdc.itap.purdue.edu 808 - Referral Site
2 http://translate.googleusercontent.com 7
3 http://www.extension.purdue.edu 6
4 http://www.edustore.purdue.edu 2
5 http://edustore.purdue.edu 2

Thank you,

Gina
Replies 3
Avatar
1
Michael
The difference could occur because of the way the program determines referrers that are shown in this table. These referrers are visitor referrers, and they are determined as referrer for the first page viewed by a visitor on your site. So if a visitor viewed any page(s) on your site, then visited the referrer site and followed a link on it to the landing page, this referrer won't be shown. However, if you include requests to the landing page only, the program thinks that it is the first page visited on the site, and shows its referrer as a visitor referrer.

It explains why in the second case number of visitors shown is higher. BTW if you analyze hits for the referrer site in the same report (e.g. if you have common logs for both sites), in most cases number of the referrers you expect should be be shown as almost zero, because the program considers all these requests to be requests from the same site. As visitors first visit pages on the referrer site, and only then follow to the landing page, referrers to the first viewed page on the referring site will be shown instead of URL with link to the landing page. If it is indeed the case, you can either use the filter as you did, or you can add the landing page to the list of tracked files in Profile Properties > Tracking to view all direct referrers to it from all visitors not depending on their first viewed pages.
Avatar
0
Gina Price
Michael,

Are you saying that the first scenario in which I did not use any include filters is capturing the visitor just once to the Landing Site A? So, if they come in, then leave, then come in again, I get 1 for the visitor referral?

If I use include filters, am I getting multiple entries from one visitor?

What would be the most accurate way to report how many people were referred from the referral site to the Landing Site A?

My "Landing Site A is a place where PDFs are saved and does not have a landing page or public facing page. The PDFs are just pushed to the Internet.

Unfortunately, my logs are not combined and I am working with different hosts.

Thank you,
Gina
Avatar
1
Michael
The best way to get all referrers for a file not depending on other requests of visitors is to add it to the tracked file list in Profile Properties > Tracking. After it you will be able to view referrers to the file in the Tracked Files report section (you need the report on referrers for the file, not report on referrers of visitors that accessed the file that is also shown in this section).

The program determines visitors by the IP addresses. If a request from an IP address came after some time (timeout) since the last request from this IP, it is considered to belong to a different visitor. The timeout is set to 30 minutes by default but you can change this value in Options > Analysis > General.

If a visitor visited a page, left the site, and then returned to it, his first referrer will be recorded as visitor referrer. If you create an include filter for a page, it will be always the first visitor page, so referrers for it will be recorded once per visitor. However, if a visitor visited this page multiple times, only one referrer per visitor will be recorded if time between requests is less than 30 minutes.