Quick question for the wise

Selected answers from the Dumb SEO Questions Facebook & G+ community.

• 
  Kuba Serafinowski: Bot traffic filtering depends on GA configuration so it can vary.And yes, this would be easy to skew, but I suppose when it stops giving you enough ROI you`ll stop paying.
• 
  Chris Boggs: I personally would watch the behavior metrics and ensure they align with other referral traffic. If abnormally low (or high) Time on Site/Pages viewed or super-high (or low) Bounce rate by comparison, I would begin to wonder. if everything aligns and the conversion rate is similar to that of other like-referrers, then no problem.
• 
  Regie Macalam: You have all the right to be suspicious about the clicks your site is getting from the advertiser. Pay per click is really tricky. Why not switch to a CPM model? Or perhaps a Pay Per Lead model. If your sticking with PPC though, try asking for more details, like hour of day, city, referring pages, medium, sources, etc. Should be easy to collect in a custom report using Google Data Studio.
• 
  Michael Martinez: Google Analytics isn`t the best platform for that kind of tracking. They don`t report keyword referral data or IP addresses, so if anyone is monitoring the traffic closely they`ll have to compare that GA data to some other source and make a guess.There are (expensive) affiliate tracking applications you can lease on the Web for this kind of tracking. You could also try to roll your own, if you have the programming skill and time.
• 
  Tim Capper: As per Michaels comment, just way to easy to manipulate if they were inclined to.Have they added specific utm tracking on the link to your site?
• 
  Frank Watson: you could use GTM to know clicks and track conversions - and as mentioned above make sure the ROAS is acceptable
• 
  Paul David: The hosting we offer comes standard with BOTH webalizer and AWstats and BOTH give you full stats and IPs including actions people take on your site... Check to see if your hosting account has this, if not, install something similar to track activity and IPs coming to your site...Then compare this data to googles.
• 
  Richard Hearne: Neither Webalizer or AWstats are fit for purpose here. I`d trust GA data over both of them. If you`re using GTM to inject GA, you could track scroll depth, dwell time and a few other metrics that would be far harder to spoof. Even pages per session, and goal conversion metrics for this cohort of traffic should help you determine if the traffic is iffy.
• 
  Michael Martinez: So, we may all have overlooked a very important question in the original post:"Wouldn’t it be technically very straight forward to automatically/artificially generate site hits for the benefit of an advertiser in this situation?"Yes, for someone who knows how to do this. And there is no analytics software that can guarantee to filter out fraudulent clicks regardless of what you do with it.Whether you are being billed for fraudulent clicks is a very difficult question to answer. Large ad networks like Google struggle to catch all fraudulent clicks. We don`t know how effective they are but we do know they invest millions of dollars every year in staff and technology to fight click fraud.So you should really ask yourself if you`re satisfied with the current business model. Regardless of how many suggestions you receive for improving the integrity of the reporting methodology, you`ll never know for sure if you`re being cheated.If you can live with the doubt then do what you can to verify the results you`re being billed for. Otherwise, find a different way to get traffic. Pay-per-lead or pay-per-sale are two alternatives to consider.