QUOTE(Michael Martinez @ May 17 2011, 06:11 PM)
Well, wait a minute. You may have wasted hours looking at the wrong data but you still haven't shown whether (or not) there are any referrals coming from the Telecom's campaign. That example log entry you posted is in a slightly different format from what I'm used to (they can be in a variety of formats) but I think it may have "slots" for referring Website and keywords.
You want to search the log for the Yahoo in the referral data. Count THOSE references. (The lines won't mention Slurp).
You're looking for PPC referrals, right? You mentioned an SEM campaign.
Michael, I have to say, I didn't waste hours looking at the wrong data, I just didn't realize the data I was looking at on the Yahoo side were bots. I have shown that the telecom have not sent referrals. On March 1 the telecom is showing 20 referrals in their reporting, there are only 2 referrals for Yahoo (of any kind) in the DS logs, March 2 telecoms = 19, DS = 0, March 3 telecoms = 21, DS = 2 and it goes on and on and on just like that.
Where the telecom got caught is that this site has done so poorly that if you take all of the traffic from any engine (either organic or paid) it doesn't come close to adding up to the numbers reported by the telecom. I'm sure that they've never had a site in such poor shape before and that's why it's so easy to see the discrepancy. I'm sure that these guys always tell their customers that their paid referrals are mixed into the organic results on GA (because they won't link the AdWords account appropriately and most people using this service don't know how to properly track other paid campaign activity via Analytics) - but when I saw that the telecom were reporting over 1,000 visits per month from their end, but the site only has 399 total visits from anywhere (referral, direct, organic & paid combined) it was clear something was way off. The first thing the telecoms said was that GA wouldn't show any results for their traffic (which is hogwash of a sort I just wouldn't swallow) so I went to the logs - but I also looked at the accounting for the site. In 2 years, they've had 24,000 visits and have had 420 conversions (1.73%), which tied beautifully to GA. If this telecom had driven 20,000 that GA hadn't tracked there would have been sales that didn't tie with GA - and it's just statistically impossible that 20,000 visitors from paid ads would never convert, while 24,000 visitors from everywhere else had. The data is clear day because the site stats are the Charlie Browneist. That's about the change though and hopefully not too late.