Counts increment of returning visitors for suspicious clicks?

suspicious clicks

[Petra Kaiser]

We notice a trop in adsense clicks and revenue in May, June and July which is almost the same amount as the percentage of total returning visitors. In our case increment of returning visitors is from 15% to 20% for the last 3 month and 10% to 20% for the last 8 months.
Only 10% of returning visitors are direct visitors but about 60% come back using our main keywords others are from several sites.
Could this be a click fraud algo to prevent from organized clicks? Has someone similar results?
(no we are not involved in any network)

[ken_b]

Interesting thought. But how would that kind of deal play out on a forum, where returning visitors might well make up the bulk of the traffic?

For what it's worth, which may well be nothing, I was just looking at a section of my site where returnees make up about 20 - 30% of the visits. Income in that section has grown in step with the traffic totals.

[jomaxx]

Personally I doubt it. Just because two numbers are about the same doesn't mean there is any connection between them.

[Petra Kaiser]

It’s not that much about returning visitors but the increment of the returning visitors, this could indicate a growing network.

[europeforvisitors]

Again, the payout percentage ("traffic acquisition costs" divided by "Google network revenues") continues to hover around 78%, with a slight rise this quarter. You can obfuscate all you want, but the fact remains that payout percentages have been remarkably consistent from quarter to quarter.

[miguelito]

sorry, don't believe in any of that, my sites are all forum based with solid core membership, 70-85 % of the clicks are from returning members and after 3 years, revenue can still fluctuate wildly and increase as well as decrease
besides, 80 % plus are on dynamic IPs so there is no pattern there to check by.

[Petra Kaiser]

Within the same time pageviews went up by 53%, this should increase revenue but revenue went down, at this moment we need 44 pageviews for a valid click, which was 28 pageviews before. The difference is just too much. In my opinion 250,000 pageviews is a good number to base a trend on.