Home > Technical & Creative Skills > Tracking Campaigns

---

😱 30-40%+ clickloss from lander to affiliate network? (8)

---

Hey guys!

Through weekend I was digging in my stats as I noticed some performance drops in my campaigns.

I went through all the traditional stuff: testing the whole funnel, landers, offers, checking for domain flags etc.

Then I started comparing clicks that I sent to network and those that they actually received.

I was shocked when I noticed 30-40% clickloss in some cases.

I'm not going to name drop for now, but yes some of these networks are very well known and they have 40% difference!

Here we are talking about clickloss from LANDING PAGE to the OFFER (not from traffic source).

So users are already on lander and going through the /click link.

To be honest I somehow never really checked these stats as I somehow trusted to my setup (Voluum tracker in the cloud, landers on VPS close to my geo's etc.).

Only thing that could cause losses is DMR (Double Meta Refresh), not sure if we have any stats on what % could be lost this way, but 40%? No way!

Another thing is, that I know some affiliate systems (Cake) do not show/count duplicate clicks and it's quite difficult to get raw click reports from their systems (hmm who knows why...)

I'm really curious what you other guys have to say to this topic, I believe many people do not focus on this stuff and it may be very easy to lost track of it.

Really, just try it and post your results here:

Go in your tracker, check how many clicks you sent to the network and then login to your affiliate network, pull a Month to day report, see how many clicks they received.

I think you will be shocked as well.

From the tech side ... Do you have any recommendations ideas what could cause these issues (beside that networks are shaving clicks?)

I removed DMR for now, thinking about replacing tracker or even networks.

I would love to hear your experience and ideas.

Btw it's push traffic.

---

I just checked some networks and direct advertisers we are using, the results are very different from one to the next. All the numbers I'm posting are banner clicks for random periods of time, to give some variety.

For example with one direct advertiser, custom tracking solution, we've sent 18500 clicks and the advertiser shows 15700, which is a loss of about 17%-18%. Voluum labeled 8% as suspicious clicks, so it's possible that the advertiser didn't count them as it was bots or proxies, which would take the loss down to about 10%.

Another one, 890 clicks sent, the network shows 1500 raw hits and 840 unique sessions. Not sure where that higher number comes from, but the uniques pretty much match. This was an affiliate network with a custom backend.

Next, affiliate network again, this one is using cake. 15200 clicks sent, network shows 13300 so a loss of 15%. 21% labeled as suspicious visits by Voluum so the loss kinda matches with that again.

...

I could go on, but the losses I see are nowhere close to what you reported. Could be due to the nature of the traffic too. Really curious about some numbers from others.

---

I had weirder stuff happen like click loss on one certain offer only on the network even when running more offers. So it likely has to do with some filters. DMR has a small impact, probably but doubt it's even close to 10%, let alone 40%.

Is this "loss" for all offers in your case?

---

Damn, Manu!

Thanks for this hint!

I'm checking it, looks like one geo, especially one lander gives me 30% suspicious clicks (thanks Matej), marked as "Fast clickers" (probably due to back button script that keeps switching between lander and offer)

Voluum doesn't show unique clicks, so it's counted multiple times.

I'm going to check the data, but good stuff, this was probably causing the biggest "drop".

---

For modern browsers, you may want to skip DMR and just use referrer policy settings .

You get the same results without the traffic losses because it doesn’t slow anything down.

https://developer.mozilla.org/en-US/...eferrer-Policy

Additional Tip: For those of you reading this who use Prosper202, the Purlink zero redirect option eliminates tracker latency related clickloss from lander to affiliate network. It also reduces click loss due to slow redirects from the network’s own tracking links

---

Yeah, I'm using this already, but still was "securing" it with DMR. Looks like though it's now supported quite good across the board https://caniuse.com/#feat=referrer-policy.

---

Yep support is much better now. You may also want to check out this from FB
https://www.facebook.com/notes/faceb...s/392382738919

---

Since we've covered this topic in VoluumLAB, let me copy what we've come up with there (so people who aren't in VoluumLAB can see as well).

Unique clicks: yes, we plan on adding that, unique visits that were added recently are first step towards those smart rotations, so we're getting there, incrementally As for unique clicks, there is a way to get them, though by counting uniqueness by IP only - you could get a report for a campaign grouped by IP, export to csv, and then filter clicks where it's more than 0 and see how many rows you have.

Clickloss: it could literally be anything - quality of traffic dropping down, the network shaving clicks, bots or the aff network counting only unique clicks, so pretty much everything that you've already covered here. DMR shouldn't be the case - if it caused ~2% of clickloss that would already be a lot, and 30% is impossible.

Since anti-fraud kit detected 30% suspicious clicks there (fast clickers), this time it was probably caused by one landing page with an aggressive back button script sending users between the lander and the offer.

---

Home > Technical & Creative Skills > Tracking Campaigns