since it comes up in every thread, lets do it properly. shaving = the network quietly under-reporting your conversions/revenue so they keep a slice. its the oldest trick and its very hard to prove, which is exactly why they do it.
how I sniff it out after getting burned enough times:
your own tracker vs their reported numbers. if you can’t track independently you are flying blind and that’s the point. get a tracker.
a sudden “quality” drop with no change to your traffic. you didn’t get worse overnight, they got greedier
conversion rate that mysteriously caps right under where a payout tier would trigger
every metric fine except the one that pays you
you rarely get a smoking gun. its a pattern. trust your gut and your tracker over their dashboard
for me the only thing that ever show me the truth about shaving is run the same campaign on a second network. all the rest is just guessing, you cannot know.
your own tracking is the only thing they can’t argue with. without your own numbers every dispute is your word against theirs, and you know whose word the network’s going to take. track it yourself or accept you’re flying blind.
the grim part is a little shaving is basically priced into the whole business. everyone assumes a bit of skim, the only real question is whether it’s a bit or a lot. trust but verify, and mostly verify.
and i’ll die on this hill. if the only numbers you have are the network’s own, you can’t detect shaving at all, you’re trusting the one party with a reason to shave to report honestly on itself. your own tracker on your own side is the only ground truth. the gap between what you counted and what they paid is the whole measurement. you can’t catch shaving with the shaver’s ruler.
splitting the same traffic across two networks at once is the cleanest practical test there is. send it evenly to both and compare what each one reports and pays for identical traffic. if one consistently counts less than the other, there’s your shave, no accusations needed, just two columns side by side. costs a bit to run traffic through a test network but it’s the cheapest insurance against getting quietly skimmed for months.
yeah, a few percent skimmed that’s hard to prove and not quite worth walking over is just the cost of dealing with a lot of networks. what you can actually catch and act on is the big shave, where they report dramatically less than your tracker counts. so track everything, let the noise-level gap you can’t prove go, and move your traffic the second the gap gets large and stays that way. you won’t catch every skimmed percent but you’ll catch the theft that matters.
same thing frank said, running the same traffic to two networks at once is the cleanest test there is. send it evenly and compare what each reports and pays for identical traffic. one that keeps underreporting is showing you the shave with no accusation needed. a test network alongside costs a little and it’s the cheapest insurance against months of quiet skimming. you need a second counter on the same traffic to see the gap.
if a bit of shaving is basically everywhere then chasing zero is a waste of time. point your detection at the networks shaving badly against that baseline instead. the real money goes to the heavy shavers, and those are the ones you can actually spot and drop.
spot on. you can’t tell if a network’s shaving without your own numbers to put next to theirs, so independent tracking is the one thing that makes it detectable at all. run your own counts, compare them to what the network credited, and the gap is the shave. without it you’re just trusting them to report honestly on their own behaviour, which is no check at all.
shaving you catch by watching your own numbers against what the network report, over time. if your tracker say 100 conversions and they pay 85 every week, is not variance, is a pattern. small networks i test with small spend first, always. trust is earned with the payout, not the pitch.
the cleanest shaving test i know is a small controlled spend where you count conversions yourself and compare to what they report, repeated over a few weeks. one week proves nothing, a consistent gap does. i never scale a new network before that check clears.
the other shaving tell nobody mentions is rebill, not just the first conversion. a network that reports your signups honestly but quietly loses your rebills over time is shaving the back end where you are not looking. watch the whole lifetime, not the front.
The clearest shaving tell for me is when their reported conversions drop while my own click quality has not changed at all. If my numbers say the traffic is steady and theirs quietly dips, something is being skimmed. You need your own postbacks, otherwise you are just trusting them to report honestly on themselves.
A community for adult-industry professionals. Independent reviews live at
xwebmasters.com.