Google Analytics underreporting referrers?

Is Google Analytics underreporting referrers from my top site?

[HistoryBuff]

Hi all. New to the group and thanks for your help.

We had some difficulty tracking referrers through WebTrends for one of our sites for a few weeks, so I decided to use Google Analytics in our report for that period while we tried to rectify the situation.

Well, we had our first full week of WebTrends referrals for that site last week, and the difference in numbers is staggering! But only for one referrer.

The top referrer on GA and WebTrends are the same for that week, but WebTrends has over 20,000 more visits recorded from that referrer than GA. The rest of the numbers are different, but not by such a large margin.

Any thoughts on why this might be the case?

Thanks in advance!
Sean

[cgrantski]

In WebTrends, "referral" can happen at any point in a visit, so a visit can have more than one referral. If people are backing out from your site and re-entering from that referring site (or going somewhere else and re-entering your site from somewhere else, within 30 minutes) then the same visit will have multiple referrals.

The important thing to notice is that WT uses the word "referral" and not "referrer."

If you want only the first referrer for the whole visit in WebTrends, you need to set it up to do so.

I think GA only counts the referrer for the first hit. It's a lot easier on the back end if GA does it that way, so it makes sense that they would take the other approach.

Whether this is underreporting or not depends on your value for those other referring events that bring a visitor back to your site once they've backed out or whatever.

[HistoryBuff]

Very enlightening! Thanks for the tip.