Home > General > Affiliate Marketing Forum

---

Blacklist method on pop traffic that works for me (16)

---

Hi, STM helped me a lot. Thank you.
Here is my method to bl positions.

1. Run a couple thousand visits.
2. Filter positions which has no conversion, and has more than 50 visits and 1/2 of average ctr.

If average ctr is 10%, then find positions has ctr lower than 5%.

It works pretty good for me.

---

Yup, it's important to define a set of rules for cutting placements, the one you chose is pretty aggressive - 50 visits sounds like a pretty low amount to me. But if it's working for you ... this just shows that any problem can be addressed in multiple ways

---

Yes, 50 is too aggressive. Today I changed it to 100 visits. I can get green pretty quick if I go aggressive cut.
I think I should create a new campaign, target to the blacklist positions and run some traffic, see if I cut wrong positions.

---

Yup, this is something that many people do ... be aggressive with the initial tests, then return to the "better" of those already paused placements and give them another go with a fine-tuned funnel. Great thinking!

---

How do you filter placements on Voluum?

---

Hmmm I might have to try something similar to this. Just started on pops recently and been cutting based solely on payout. Which itself is not a bad rule, but it's eating my ass up right now!

---

You have to pass the placements to Voluum first, by using the dynamic tokens of the particular networks you are using. Then you can drill down into your reports at voluum and analyze your data by placements. Then you go back to your source and block placements that are not working for you.

---

Yesterday I started a campaign, target only positions in my blacklist. Give it a very low cpm(1/3 of normal). I got +50% ROI.

---

I do something similar but I have a question about higher ctr. Lets say my avg ctr is 5%, at what %age above 5% would you cut? like now i have a placement not converting for like a few hundreds visits at 13% ctr with 0 conversion. Cut it?

---

I don't think high ctr is a problem. Does your lander redirect to offer URL automatically after a few seconds?

---

this lander nope..

So keep all the high ctr ids, and cut the obvious crappy ones? (already had 1000%+ on some ids, I cut them fast though)

---

Hey dollar, how much increase in the ROI you often experience after cutting placements?

---

Awesome stuff, I would find that pretty aggro aswell.

I had conversions in placements with CTR around 5% and a low amount of visits.
Everybode has their own approach tho and it for sure depends a lot on TS and GEO aswell and
your comfort level in that vertical.

However I am currently using 100-200 visits with CTR <5%.

Cheers.

---

I almost never cut on CTR anymore, because there's a lot of factor, i.e. a placement being visited mainly by mobile or by a demographic with newer phones or whatever. Let's say that your average CTR is ±50% if you run like an aggressive AV lander, not counting back button redirect or anything. Some bots might click a lot, some less. Some get trapped in JS on the site and screws with the CTR ... but you'll still see conversions on those, and they might even convert better than others.

The only times I'd cut on CTR would be if you have like 100 visits and 0 clicks, then I'd blacklist it. Mainly on Wifi. So if I run wifi in large geos where traffic is plenty, it's fine.

With a good offer you can get away with a lot anyways, so I like to let each segment spend enough and placements at least 2x payout, or 3-4-5 on the low paying 1 click flows on carriers, where you bid higher and convert a higher %.


Edit, if you're worried about bots -- most TS have more sophisticated bot detection than simple CTR rules of thumb, and you can often see the ISP in your affiliate network or in Voluum. Anything from data centers or where you see one IP/range acting weird. I believe that's more accurate.

---

This is great, very helpful. Thank you.

---

Cut using slack rules and you would waste money.

Cut aggressive and you would be cutting out potentially profitable placements.

So - each of us will need to find what we feel is the sweet spot.

I think that cutting at 50-1000 impressions is way too stringent by any standard though. Wouldn't be a bad start if you're trying to identify an initial set of the best placements to do testing with, in order to save on test budget. But once you have a good offer+lander combo, be sure to retest at least the bigger placements that got blacklisted from before.

As for cutting based on CTR - if a placement has 0% CTR or unusually low CTR compared to the majority of placements, that would be good grounds for blacklisting. Also, when you have a few conversions already, you can identify a "normal" range of CTR, and cut placements that have CTRs that fall outside of that range. Again, it boils down to how aggressively you want/need to cut.

As for bot placements - if running wifi then bots can be a real problem. There's caurmen's script, and there's also afflow stats. You can siphon a percentage of the traffic to afflow to identify high-percentage bot placements and cut them that way - learned that from zeno during the 6WAMC.

Another major consideration, is how much traffic volume you have available. If you're targeting BH Viva, cut sparingly or you'd have very little left over afterwards. On the other hand if you're targeting ZA MTN, cutting aggressively would be a necessity.



Amy

---

Home > General > Affiliate Marketing Forum