Home > Technical & Creative Skills > Tracking Campaigns

Click-Loss and Tracker Speeds. Need Help. (9)

05-14-2019 05:36 AM #1 chris r (Member)
Click-Loss and Tracker Speeds. Need Help.

Hey guys,

I wrote this in my 40 day follow along, but it seems like not too many eyes end up over there, so I figured I'd try my luck here too. I apologize if this is frowned upon.

-Chris

Originally Posted by chris r

Update:

I did a bunch of tracker testing since my last update. I don't have any screen shots (I can take some if need be) but the results are pretty interesting.

I tested Voluum, Binom, Kintura and Bemob. All four tests received traffic from Propeller (OnClick Popunder) and used the same offer that was for Australia.

First up is Voluum. With this setup, I used Voluum's unique dedicated domain, DNS, and SSL cert. I would have preferred to use my own domain and DNS through Cloudflare, but they do not offer a free SSL cert with custom domains on their entry level plan.

Propeller showed 20,665 impressions. Voluum showed 7,254. Click-loss of 64%

Next up, Binom. With this setup, I used a Digital Ocean VPS (located in Frankfurt), a domain from NameCheap, DNS through Cloudflare and SSL through Binom.

Propeller showed 26,867 impressions. Binom showed 7,594. Click loss of 71%

Third was Kintura. With this setup, I used a domain from NameCheap, DNS and SSL through Kintura.

Propeller showed 23,886 impressions. Kintura showed 9,454. Click loss of 60%

Last one, Bemob. With this setup, I used a domain from NameCheap, DNS through Cloudflare and SSL through Bemob.

Propeller showed 21,456. Bemob showed 8,573. Click loss of 60%

I also tested each trackers campaign link on GTmetrix, using the same exact landing page, five times, in four different countries. The landing page is hosted on Amazon S3, CDN and DNS are Cloudflare.

Here are the results:

Australia

1.6 / 0.8 / 0.8 / 1.0 / 0.8 - NameCheap domain (No tracker)

2.6 / 1.7 / 1.8 / 1.6 / 1.6 - Voluum

2.1 / 2.1 / 2.4 / 2.1 / 2.1 - Binom

0.7 / 238ms / 0.5 / 284ms / 406ms - Kintura

2.3 / 1.9 / 1.9 / 1.9 / 1.9 - Bemob

UK

0.5 / 201ms / 186ms / 208ms / 151ms - NameCheap domain (No tracker)

1.1 / 0.6 / 0.6 / 494ms / 0.6 - Voluum

482ms / 0.6 / 456ms / 438ms / 0.6 - Binom

0.6 / 259ms / 283ms / 282ms / 344ms - Kintura

0.6 / 432ms / 324ms / 422ms / 332ms - Bemob

China

1.1 / 1.4 / 355ms / 0.6 / 0.6 - NameCheap domain (No tracker)

1.5 / 1.2 / 0.9 / 1.3 / 1.2 - Voluum

1.5 / 1.4 / 1.8 / 1.5 / 1.4 - Binom

1.6 / 0.9 / 0.8 / 0.8 / 1.1 - Kintura

0.9 / 492ms / 481ms / 0.6 / 0.8 - Bemob

US

346ms / 271ms / 0.7 / 299ms / 0.5 - NameCheap domain (No tracker)

1.0 / 1.3 / 0.7 / 0.6 / 467ms - Voluum

0.8 / 0.8 / 0.9 / 0.8 / 0.7 - Binom

0.5 / 431ms / 366ms / 427ms / 391ms - Kintura

479ms / 422ms / 377ms / 373ms / 496ms - Bemob

-Chris

05-14-2019 09:32 AM #2 matuloo (Legendary Moderator)

Interesting numbers for sure. 30% clickloss for POPs isn't that bad, especially for some GEOs. But 70+% is way to much.

May I ask how you conducted this test, because the difference between bemob and the others is very big. Especially when it comes to clickloss, the loading time differences were not that big.

Were these run at the same time? Or you started one, then paused, started another one? Were the bids the same for all campaigns? Targeting, WL/BL ... everything the same?

05-14-2019 12:27 PM #3 Mr Baffoe (Veteran Member)

Hi interesting data can you clarify the flow is it:

1. Ad -> tracker link -> your landing page on your domain
2. Ad -> tracker link -> offer landing page
3. Ad -> your landing page

Options 1-2 will cause more traffic loss than option 3.

The ideal setup is option 3, this uses a JavaScript tracking pixel on the landing page. Your lp should also be on a cdn for best results.

05-14-2019 02:51 PM #4 voluum (Veteran Member)

Apart from the factors that were already mentioned here, traffic quality might also be relevant. It might be that Voluum didn't count a number of impressions Propeller did on purpose, if these were bots that we know of and have in our libraries, Voluum might have filtered them out. I'm not saying that's the case here but I have ways of checking it and I want to look into it further. I'll send you a PM for the campaign details and check the data.

Karolina

05-14-2019 05:23 PM #5 chris r (Member)

Originally Posted by matuloo

Okay that's good to know. I thought 30% was way too high as well.

I don't know if it was the best way to test, but I just setup the same exact campaign in the three different trackers, ran them on propeller and recorded the results. I stopped the Voluum one a little early because it was looking so bad.

They were ran at different times. The bids, targeting and everything else was exactly the same.

Originally Posted by Mr Baffoe

Hello. It would be option 1.

My landers are hosted on Amazon S3 but I do use Cloudflare for a CDN. I was using Cloudfront before, but for whatever reason it never seemed to perform quite right.

That's interesting about option 3. I suppose this would cut down on all of the redirects right?

Originally Posted by Voluum

Oh okay. I didn't even know you guys did that.

I'll reply back to ya right now. I appreciate you looking into this for me.

-Chris

05-14-2019 10:20 PM #6 matuloo (Legendary Moderator)

They were ran at different times. The bids, targeting and everything else was exactly the same.

Would be interesting to run them all at the same time, or at least give each of them the same amount of time, for example a whole day. I know these tests are hard to run and it costs $ too. But the ideal case would be to route this through one additional redirect, that would distribute the traffic evenly across all the 3 variations. Such numbers would be more representative as it's quite normal to see different performance throughout the day and it also depends on what particular placements send traffic at any given time.

05-16-2019 06:02 PM #7 chris r (Member)

Originally Posted by matuloo

Yeah that probably would have been the right way to do it. At the same time, I wasn't really trying to run an official test. I just noticed this issue in one tracker, then decided to sign up for the "industry standard" and then compared everything to what I was using.

After talking to a few representatives from a few trackers. It's sounding more and more like the traffic was just shit. I may run a few more small tests today with PopAds aiming at the same country. It'll be interesting if the results are similar.

-Chris

05-16-2019 07:25 PM #8 matuloo (Legendary Moderator)

After talking to a few representatives from a few trackers. It's sounding more and more like the traffic was just shit.

Yup, that's an option too. Check this thread : https://stmforum.com/forum/showthrea...iliate-network Erik was experiencing a high % of lost clicks and in the end he figured out it was caused by poor traffic too.

05-17-2019 05:41 PM #9 chris r (Member)

Originally Posted by matuloo

Yeah it's looking like this might be the issue. I'll definitely give that thread a look.

Thanks dude!

-Chris

Home > Technical & Creative Skills > Tracking Campaigns