Home > Paid Traffic Sources > POP / PPV / Redirect

---

How Not To Get Kicked Out Of Offers When Using Pop Traffic (22)

---

Is there a way for you not to get kicked out of offers when promoting using pop traffic? When I start getting profitable and start to scale and drive more leads, I get paused due to bad quality. This has happened a few times already.

I know that pop traffic is the worst type of traffic but are there offers that you can run long term without getting paused? What vertical would match this traffic type. I was promoting adult dating offers btw.

How do you drive pop traffic to offers sustainably?

---

Is pop traffic allowed for these offers?

If yes, I’d suggest passing placement IDs to the network, then work with your AM to identify poor quality placements, so that you can keep running those offers.

---

It´s not the worst type of traffic

Anyway, from my experience adult dating can be a bit tricky on Pops because adult advertisers can be more strict about the quality compared to others.

But here are a few tips to increase quality on Pops.

What you definitely should do is to "clean" your traffic a bit so that you try to only send real users and matching traffic to the offers.

This starts by using language targeting, only get traffic from users in the native language(s) and maybe english.
When your trafficsource doesn´t support language targeting do it with rules in your tracker.

And there we are at the next point - Redirect rules in the tracker.

Use rules in your tracker to get rid of traffic that doesn´t meet the offer requirements and that looks suspicious.

This helps to increase your quality but it can also help you to make some money from traffic that wouldn´t be able to convert on our offer anyway.
It doesn´t matter where you buy traffic or what you run, a part of your traffic is always something you don´t target.

So here are some examples for rules you can use:

- If Country is not "Country you target" send the traffic to a Smartlink or a different other offer that converts on that Geo
- If traffic is proxy send it to smartlink
- If traffic is not the OS you target send it to smartlink
- If traffic is not the device type you target send it to smartlink
- If traffic comes from a server or datacenter send it to smartlink
- If traffic has a bad user agent send it to smartlink
- When connection is corporate, unknown or NetSec send it to a smartlink

These are only few examples but that alone can already help to increase the quality of your traffic.

Then you can do what platinum said and send placements to the networks.
When you do it then of course you risk to expose important data of your campaign so I would do it with all advertisers or networks.
But that´s the only way to know if a placement is really good or not.
Just because a placement is converting very good for you it doesn´t mean that it also makes money for the advertiser.
So you can only know it when you run your stuff that transparent.
When you can trust them it´s a very good way to increase the quality alot and this can be worth it when you then get higher payouts and/or higher caps.

But what I always do is to send trafficsource ID in (mostly) s1 parameter and campaign ID in s3 or s4.
That way you don´t expose too much info but still get an idea what works and what not.

And last but not least it also depends what for dating offers you run.

SOI = more risk to run into quality issues
Rev Share = not that high risk to run into quality issues because there the risk is on your side

---

POPs are quite known for poor lead quality in adult dating... that's a fact.

The people that I know run dating on POPs are doing 2 things, at least large part of them:

- what twinaxe suggested, so cleaning the traffic and getting rid of the segments that are known for quality issues ... VPNs, language not matching the GEO, BOTs... this all has to be removed/blocked.
- mixing POPs with some other traffic type. So for example, they would optimize the traffic to reach max lead quality with banner traffic and then mix in some POPs to get higher profits.

And yes, passing the source/placement IDs can help you identify the worst placements and get rid of those.

---

Thanks @twinaxe and @matuloo
Do you know any tutorial or step by step guide on how to do the passing of placement ID to the affiliate network? Can you point to the right direction? Thank you for the tips.

---

Just the same way you pass the click ID.

When you send the click ID in the offer URL in the s2 parameter like &s2=CLICKID send the placement ID for example in s3 parameter like &s3=PLACEMENTID

Better don't use the s1 parameter for it, several CPA networks use s1 for sub sources and will the parameter for you when you send too many different values there.

---

To add to this, some advertisers will ask for a specific token to be used for placement ID passing, so ask your AM what token to use for this. In most cases the default "s3, s4 ..." should do, but I've also seen different setups like "k1, k2 ... or pid" and a few more different ones.

---

Thanks, never thought it was as simple as that.

---

This is brilliant and I feel dumb for not setting these rules up until now.

---

There is also a downside to this The traffic that you redirect away or block, would have made some leads at the original offer and you will (almost with 100% certainty) make less from the smartlink. So many affiliates refuse to do this, it's hard to accept the initial loss. But in the end, the higher quality "should" pay for it.

---

Oh for sure. Smartlinks make way less than the original offer. But. They also make way more than 0, which is what a lot of garbage traffic will make anyway.

And for traffic that seeps through even when you try to block it on the source (Android Webview) it's a great way to redirect garbage traffic to a smart link which has a higher chance of converting, even if it's only a few cents.

It's also a great way to add some friction to spying by getting rid of proxy traffic (you'll never beat AdPlexity though ).

---

Yup, can only agree with everything you posted

---

Some awesome suggestions in this thread!

Like @matuloo has mentioned above, it might be difficult to profit, when such strict rules are applied in your tracker, especially when you are just starting out, trying to figure out which offer/landing pages/zones work best.

If you generate enough leads for your affiliate network, consider asking them for a second (or third, or fourth ) account that will have a different affiliateid. You can use it to test offers and then once your winning funnel is figured out, duplicate the campaign with the offer from your main account and setup quality increasing rules there.

This way you might stay on the offer just a little longer.

---

Absolutely correct, a smartlink will mostly make less than a single offer.

But it´s a difference to set up rules for traffic that could convert on the original offer or to have rules for traffic that can´t convert on the offer anyway.
As I said, a part of the traffic will always be different from what you target and you will only make less money when you send traffic that could convert to a smartlink instead of the original offer.

When you create rules for traffic that doesn´t meet the offer requirements anyway you can only win.
So when you set up rules for such traffic and send it to a smartlink you can try to monetize that traffic as well.

When the offer only accepts traffic from US but you also receive traffic from CA, DE, UK and AU send the other traffic to a smartlink.
When the offer only accepts iOS but you receive a bit Android traffic send Android users to a smartlink.

Also when I have rules for datacenters, servers, bad user agents and stuff then I can lose some conversions maybe but every advertiser will be happy when he doesn´t receive conversions from cogent, easyhost, leaseweb and so on

---

Sorry, I'm feeling a bit dumb here. What's all that? I want to avoid that too, I think. Where can I find out about that?

---

It's users browsing through VPNs, for example. So people connecting from different countries that block whatever type of content... frequently used to browse adult sites from countries where such content is prohibited.

Then there are visits coming from hosting companies, like digitalocean, vultr ... these are usually some scripts/bots that run on these hosts, could be some content crawlers, uptime monitors etc... one way or another, these won't convert.

There are more cases like this, but I'm sure you get the idea. To block these, you can start by looking at the ref data in your tracker but to really catch all of them, you will need an external DB that can identify these.

---

I get the VPN users, blocking proxy traffic should take care of most of them. Bad user agents should be taken care of by limiting by browsers mostly.

I'm more interested in what cogent, easyhost, leaseweb are and where I can find some info on those types of things?

---

Cogent, easyhost, leasweb and so on are datacenters, server hosting and stuff.

Basically traffic where chances are good that these are no real users.

I started collecting such stuff myself in 2011 or 2012 or so when I was working on my cloaker.
At this time I collected it all in separate text files.
Then I cleaned the data, sorted it and put it in PHP files for the cloaker.

Nowadays it's much easier to do so with a tracker.
Usually all trackers should be able to give you that data.
Back in the days I had pretty special needs so that it was better for me to do it my way

You can do it as well when you go through your tracker stats and check the ISPs.

Then you already see some that you better block/redirect.

You will also find the legit ones, the real ISPs for normal broadband and stuff.

When you are not sure about some of them just Google them.

That way you can build a blacklist pretty good but there are always popping more of them up from time to time.
Just like a BL of placements.
When you block bad placements there can still appear new ones.

So you can also collect a WL of "real" and clean ISPs.
But there the afford to do so can be very different depending on the geo.

In some Geos its pretty easy to make such a list because there are just a few big providers.
In other Geos it can be more complicated because there are so many different small providers.

---

In general, some verticals have a lower back-end performance on pop traffic, such as dating and casino offers. POP traffic is an “in your face” style. This means there is large audience exposure, however, it is not target, so the offer gets shown to people that may not be interested. Offers that rely on active users (e.g., spending users in the case of Casino and Dating) are not a good match with POP traffic. For example, the visitors aren’t interested in gambling and only testing their luck on some free spins. On the other hand, we see sweepstake offers (SOI, PINs and trials) do deliver decent quality on POP traffic, because these offers convert on a ‘broad’ and passive audience and require, only once, minimal input by the user.

---

Very good explanation, simple and straight to the point.

---

Hi everyone!
Golden info in this post!! thanks a lot!!!

This is suggestion i used to stop bots from spamming comments on one of our wordpress sites and it worked like a charm.
Maybe can be implemented on our landing pages as well, plus everyone's sugessions above about filtering users that don't match your targeting, should improve your quality by a lot.

The solution is quite simple when you think about it.
The usual neighberhood friendly bot wants to run forwared until it cant no more. He stops to fill up forms, say hi to the cuties, and jump from page to page via links until he completed his mission.
Than he returns back home and wait his turn in the queue until he can start running to a different page.

So we need to trick him to let him think he is done, or at least redirect him to some fun dark road it can never get out of. (infinite loop redirect chain)

To solve this in wordpress, we added an additional input to the submission form and hid it.
Most bots, see only code and don't really care much about CSS and design.

So mr. bot gets to your page, fills up the form and submit it so it could get to the next page, and continue running (By wordpress he just happy he completed his job and went back home)
Your backend server gets his submission, test for the label to see if its check-marked, or filled up with anything, and then redirect the bot to a deserted island for a timeout.
If its a human, it lets it pass and arrive at the target (for wordpress, the thank you page)

Now, a couple of additional thoughts:

Then you only left with the bots that went to college with some smart alogo we didn't predict or thought of, and very complex logic, but there aren't too many of them.

Something like this, plus target filters, and additional elimination, should leave you only with human traffic, that you are actually looking for.

(PS: credit to some cool dude I found on google who suggested this solution, can't find the post now, but I am sure it's still out there if you look really well)

Work hard, play hard, make money!
Yossi

---

Hehe, the good ol' college bots with PhD.

Thanks for posting the tip.

Yes, most bots are not really meant to be highly professional.
It's mostly enough when they emulate some basic human like behavior so a clever trap can help to get rid of them.

---

Home > Paid Traffic Sources > POP / PPV / Redirect