Home > Paid Traffic Sources > POP / PPV / Redirect

CHEATSHEET: DO's and DON'Ts of Pop Traffic (54)


03-10-2016 06:55 AM #1 vortex (Senior Moderator)
CHEATSHEET: DO's and DON'Ts of Pop Traffic

Pop is a popular traffic type for newbies and pros alike. Below are some do's and don'ts that, when used appropriately, can help you maximize profits and minimize losses.


********************


DO's:

DO test multiple bids from the start. 3-5 camps at staggered bids to see which one give the best ROI.

DO use a self-hosted tracker if doing volume. A hosted-for-you tracker that charge by events can be significantly more expensive than a self-hosted tracker when you start doing volume on pop.

DO be aggressive if allowed. Check with both your affiliate network and your traffic source first to make sure you're not breaking rules.

DO test low-competition geos first. Geos with lower average bids and fewer advertisers. Asia, Africa, & LATAM America are gold!

DO run broad-appeal offers. Important as pop traffic is non-targeted.

DO consider testing multiple offers from the start to increase your chances of finding a good one. Also test same offer on different networks, and different offer landers for the same offer.

DO get rid of bad placements. See here for a helpful tool.

DO start your own global blacklist. Some placements just won't convert! Consider adding them to your global blacklist and apply it to future campaigns (thanks kepe95).

DO optimize landers for loading speed. Use a CDN and optimize lander code.

DO test back button and popunder scripts. Split-test direct to the offer / scarcity popup on same lander / a new lander and offer entirely (thanks to dazed1).

DO test popular landers that look different, to establish benchmark stats. Then beat them with your custom landers. Do some manual spying (thanks to simon_89), or a spy tool.

DO make landers attention-grabbing. Turn the visitor's focus from what they were doing to your lander and offer.

DO test other available ad formats. e.g. pop & redirect traffic at zeropark; popunders/popups/tabunders/tabups at popads.

DO verify lead quality before scaling. Especially important for pop traffic!

DO actively ask for pay bumps. A natural thing to ask following a lead quality check.

DO use traffic network tracking tokens to help you optimize your campaign. You can find tracking tokens here (thanks to mitesh muley).


********************


DON'Ts:

DON'T scrimp on infrastructure! A shared 10$ server won't fit the bill if 50k impressions per hour are hitting (thanks sebastian_r!)

DON'T be afraid to test high bids, because you may get higher conversion rates as well. Set a low budget!

DON'T start mass-cutting without improving your funnel first. When a good chunk of your traffic is profitable, cut anything that doesn't give you the ROI you want.

DON'T cut anything without drilling down into data. See this post and this post.

DON'T dismiss low-payout offers. They can make good money if the target geo has cheap traffic prices that convert well.

DON'T turn off your frequency cap, at least in the beginning. 1/24 hours for example is a good start.

DON'T be cheap when it comes to getting proper translations! Use OHT for speed or Fiverr for cheap pricing.


********************


Pop Sources to Try:

Mainstream pop networks:



Networks with adult pop traffic:


********************


Useful Pop-Related Posts:

Ruby Tune's Step-By-Step Pops Tutorial

MrGreen: Most-blacklisted placements on PopAds

sebastian_r: various great tips on pop

kepe95: various great tips on pop

dazed1: various great tips on pop

Back button redirection - where to redirect?

Anyone made decent money with pop traffic?




Amy


03-10-2016 12:47 PM #2 vortex (Senior Moderator)

Here's a more newbie-friendly version of the above, with certain points explained in greater detail.


1. Popular mainstream pop networks: Zeropark, PopAds, PopCash, Adcash, PropellerAds, WigetMedia, Gunggo, Mediahub, SelfAdvertiser.

Networks with adult pop traffic: Zeropark, PopAds, Exoclick, JuicyAds, Ero-Advertising, Traffic Junky, Traffic Force, Traffic Factory, AdamoAds.



2. Run broad-appeal offers. Since pop traffic is non-targeted for the most part (i.e. doesn't allow targeting by user demographic), make sure your offer will appeal to the general populace.



3. Don't dismiss low-payout offers. Pop traffic can be purchased at relatively cheap rates (compared to other traffic types). Even offers with $0.1x payouts can make you good money, if the target geo has cheap traffic prices that convert well.



4. Check how aggressive you can get. Most of your competitors will be running aggressive, so you need to do the same if possible.

Check with your affiliate network on how aggressive you can get when pushing the offer. Also check your traffic source's creative requirements to confirm your landers are in compliance.



5. Test low-competition geos first. Some geos are easier to make profitable than others. If you want to see green faster, try geos that have lower average bids. You can find this information in the traffic source's inventory report or traffic estimator. Alternatively you can ask their support.

Also make sure the traffic source has enough traffic in that geo, for your targeting options, to be worth your time.

As well, some sources will also tell you how many advertisers are running in each geo (e.g. PopAds), which is another way to gauge competition levels.

Where to look? Asia, Africa, & LATAM America



6. Test popular landers that look different, to establish benchmark stats. When testing a new vertical on pop, with have no proven landers, test 5-10 of the most popular landers to get an idea on what range of CTRs and CRs you can expect. Of course this will depend on the offer as well, so testing multiple offers from the start would be wise too.

How to find popular landers? Do some manual spying (thank you Simon!), or use a spy tool like Adplexity.



7. Make landers attention-grabbing. Because pops are a form of "interruption marketing", you need to use every trick in the book to turn the visitor's focus from what they were doing originally, to your lander, and then hopefully to your offer.

If the offer and the traffic source allow it, use aggressive landers, and test the common CTR/CR-boosting tricks: entry/exit alerts, backbutton redirect, audio, vibrate, etc.

Optimize page load speed by using a CDN and by optimizing the coding.



8. Test multiple bids from the start. Some traffic sources will provide an "average" or "recommended" bid, but bidding at the average will not always give you the best value for your money.

Make copies of the new campaign so you have 3-5, and set a different bid for each ranging from low to average to high, set a low budget for each (10-20x payout should be enough). Run traffic, see which bid gives the best ROI, and use that camp to test your landers and offers.

Testing different bids will allow you to a)gauge promise of offers and landers, and b)save you money by testing at a bid that give good ROI.

There's no set method for selecting bids, but see here for a sample approach for popads.



9. Don't be afraid to test high bids. On many pop sources, higher bids can get you traffic that converts better. I've had initial test camps that performed vastly different at high and low bids, where the camp was profitable at the high bid but in great loss at a lower bid. If I hadn't tested the high bid, I would have written the campaign off.

Important: Remember to set a low budget as a high bid doesn't always translate into high traffic quality!



10. Get rid of bad placements. Pop sources typically have a ton of publishers/placements, and they don't all convert the same (some don't even cost the same). On certain pop sources where bot traffic is rampant, you may need to pause quite a few placements to see green. See here for a tool to help identify bad placements.

Before cutting placements en masse though, please see the next point...



11. Improve your funnel first, cut second. When evaluating a placement, remember that even the best placement won't do well with a bad offer and lander. The same applies to other targeting, e.g. OSs, devices, carriers, browsers, time of day. So split-test offers and landers until you find a combination that can make a good chunk of your traffic profitable, and THEN cut anything that isn't likely to give you the ROI you want.



12. Set a frequency cap, at least in the beginning. More often than not, turning off frequency will bring down conversion rate. Before your camp reaches profitability, it would be wise to set a frequency cap to e.g. 1 in 24 hours.

Once the camp is profitable and doing good ROI, test higher frequencies, or even turn off frequency altogether. The conversion rate may drop, but if your overall profit amount is higher than before, then that's all you'd care.



13. Test other available ad formats. For example, zeropark has both pop and redirect traffic, and popads have popunders, popups, tabunders and tabups. If you have an offer & lander that convert, try the other formats on the same traffic source.



14. Use traffic network tracking tokens to help you track the performance of traffic segments you can target at the source. This will let you optimize for traffic segments that are giving you the most traffic and the highest ROI. You can find tracking tokens for many networks here: TrackingTokens.com.






Amy


03-10-2016 12:56 PM #3 sushiparlour (Member)

Thanks Vortex, made me think if I was getting most of those checkboxes ticked :P


03-10-2016 01:18 PM #4 sebastian_r (Member)

Nice write up!

Here's my take:

Speed, Speed, Speed.

Invest into the right infrastructure before running pops. A shared 10$ server won't fit the bill if 50k impressions per hour are hitting. I have an DVS 12GB from Beyond Hosting and use Cloudflare for DNS. The combo was able to handle x,xxx,xxx impressions per day.

Optimize Lander speed. Should not be bigger than 100-300kb. The quicker the page loads, the higher your chances to convert. The difference between .1 and .15 in CV can already make or break an pop camp.

Non aggressive won't fly. I've spent $xxx,xxx on pops and did not have an single non aggressive camp working.

The pop traffic sources are all stealing from you (I consider 50% bot traffic stealing). Cutting like a madmen is the name of the game. 250 impressions without click, 1000 impressions without conversion for low payouts (stretched throughout the day). With high payout pins it can be a bit more.

If you hit something hard, most networks or their employees will copy your camp. Cloak.

Antivir, iphone, whatsapp, and utility apps are doing volume. The rest is a side show.


03-10-2016 01:56 PM #5 vortex (Senior Moderator)

Quote Originally Posted by sebastian_r View Post
Nice write up!

Here's my take:

Speed, Speed, Speed.

Invest into the right infrastructure before running pops. A shared 10$ server won't fit the bill if 50k impressions per hour are hitting. I have an DVS 12GB from Beyond Hosting and use Cloudflare for DNS. The combo was able to handle x,xxx,xxx impressions per day.

Optimize Lander speed. Should not be bigger than 100-300kb. The quicker the page loads, the higher your chances to convert. The difference between .1 and .15 in CV can already make or break an pop camp.

Non aggressive won't fly. I've spent $xxx,xxx on pops and did not have an single non aggressive camp working.

The pop traffic sources are all stealing from you (I consider 50% bot traffic stealing). Cutting like a madmen is the name of the game. 250 impressions without click, 1000 impressions without conversion for low payouts (stretched throughout the day). With high payout pins it can be a bit more.

If you hit something hard, most networks or their employees will copy your camp. Cloak.

Antivir, iphone, whatsapp, and utility apps are doing volume. The rest is a side show.

Wow thanks sebastian! So nice to have an expert on pop traffic chime in to make the list better!

I've incorporate some of your point into the list. Was debating on whether to talk about lander speed because it's not exclusive to pop sources, but I guess as long as it's relevant and helpful that all that counts.

I can't openly recommend cloaking but will include a link to this post so people can see that suggestion. Knowing a wider range of possibilities is always good!

Your observations on niches are solid - although at the moment I'm "putting on a side show" by pushing something different. Only low 4-figures/day at the moment, but can turn into mid 4-figures or 5 when fully scaled.

Your advice is much appreciated - thank you very much.


Amy


03-10-2016 02:47 PM #6 sebastian_r (Member)

Quote Originally Posted by vortex View Post

Your observations on niches are solid - although at the moment I'm "putting on a side show" by pushing something different. Only low 4-figures/day at the moment, but can turn into mid 4-figures or 5 when fully scaled.

Amy
Nice!

I'm expecting the landscape to change a bit in 2016, many pop sources and advertisers (offer owners) are getting stricter. The pressure for deceptive/misleading ads is building up constantly (think carriers, google, government regulation). New offers / niches could have a good chance to take over some of the traffic share.


03-10-2016 02:57 PM #7 kepe95 (Moderator)

Spot on! Great tips vortex and sebastian_r

Another good traffic source for pops/redirects is Plugrush. It's adult, but if you've ever researched your pops pubs, you'll realize the majority of pops traffic across the board is adult, followed by torrent and similar sites.

Concerning blacklisting placements, build a global blacklist that you apply for every new campaign, you'll soon have a big advantage right out of the gate. then also blacklist individually for every country.

Imo it has also been a bit harder to make good profit from pops since new spytools have made everything pretty transparent. It's pretty easy to spy on pops ...


03-10-2016 03:41 PM #8 shishev (Moderator)

I haven't run as many pop campaigns as you guys but the "Don't"s really hit home for me. Especially the part about investing into proper infrastructure, I had to learn that the hard way with my cheapskate low end VPSs.


03-10-2016 03:45 PM #9 vortex (Senior Moderator)

Quote Originally Posted by kepe95 View Post
Spot on! Great tips vortex and sebastian_r

Another good traffic source for pops/redirects is Plugrush. It's adult, but if you've ever researched your pops pubs, you'll realize the majority of pops traffic across the board is adult, followed by torrent and similar sites.

Concerning blacklisting placements, build a global blacklist that you apply for every new campaign, you'll soon have a big advantage right out of the gate. then also blacklist individually for every country.

Imo it has also been a bit harder to make good profit from pops since new spytools have made everything pretty transparent. It's pretty easy to spy on pops ...
Thanks Kevin! Of course - how could I have forgotten Plugrush? Added!

Regarding the global blacklist - if you're talking about bot traffic, then yes, you can apply them to any geo. However, I've found that traffic from the same placement, when running different geos or even offer types, can perform vastly different. (EDIT: Added this great tip to the list!)

Definitely agree with you on it being harder to profit from pops. Still a simple way to "scalp" profits though, and certainly still one of the easiest traffic types for newbies - anyone can rip a few landers and start learning how to optimize a campaign.

Thank you so much for your input!



Amy


03-10-2016 04:24 PM #10 sebastian_r (Member)

Quote Originally Posted by vortex View Post
Regarding the global blacklist - if you're talking about bot traffic, then yes, you can apply them to any geo. However, I've found that traffic from the same placement, when running different geos or even offer types, can perform vastly different.

Another thought for the guys hitting on pin submits: different placements show different performance on carrier and wifi. I had countless placements which would bleed me out on wifi, but carrier only performed very well. Seems that the carrier bots are still an developing species. But they're certainly catching up.


03-10-2016 05:12 PM #11 vortex (Senior Moderator)

Quote Originally Posted by sebastian_r View Post
Another thought for the guys hitting on pin submits: different placements show different performance on carrier and wifi. I had countless placements which would bleed me out on wifi, but carrier only performed very well. Seems that the carrier bots are still an developing species. But they're certainly catching up.
That's a very important point that can't be stressed enough! Adding it to the list!

Campaign variables often CAN'T be considered individually. What you said about wifi vs. carrier traffic for the same placement is a perfect example. Another example: OfferX is doing 100% ROI with Carrier1 but -80% with Carrier2. If we just look at offer stats alone, we may decide to cut OfferX and throw out the baby with the bath water!

On the other hand, we obviously can't analyze every combination of all variables. That would be impractical, and would take a huge budget for anything to reach stat sig.


A "middle-ground" which I've found to work well for me, is to:

1)Identify the best traffic segments that have sufficient traffic (e.g. Android & IOS, Wifi & Carrier1).

2)Optimize offers and landers for each major traffic segment separately.

It's in my plans to do a detailed write-up on this very soon.



Amy


03-10-2016 05:34 PM #12 Mr Green (Administrator)

You might want to add this to the sheet too!

http://stmforum.com/forum/showthread...-net-Blacklist


03-10-2016 07:45 PM #13 dazed1 (Member)

Some very good advice in this thread so far!

Sebastian_r knows his stuff, he absolutely worked his ass off in 2015 to make a success from POP traffic. Always enjoyed the daily banter on Skype to this day - genuine super affiliate material and nice dude too.

I've also thrown a fair chunk of change at this traffic type, again in the xxx,xxx range across a few hundred campaigns in total. The were a few major insights for me that got me into the green and earning a nice ROI -

The first was buying my traffic in bulk without specifying wifi or carrier only. Once you start to specify or over-optimise ANYTHING in terms of targeting, you will find traffic quickly dwindles to nothing more than an expensive trickle! It's far better to let your tracker do the optimisation and send certain portions of traffic to different offers which match the visitor type. As Seb pointed out earlier, carrier traffic is where the action is at - use your highest paying and strongest converting pin-submit offer here and make the most of it! (Look for one-click flows) It's worth noting that if you can't find a good fit for wifi traffic then consider passing it on to a network such as Monetizer or YTZ.com

Secondly, watch those bid amounts - contrary to the general consensus, for me this game is ALL about volume over absolute quality - doubling your bid does not always mean you get double the impressions - start with a low bid and gradually increase until you find the sweetspot - the point where increasing the bid further does not drastically increase impressions per hour. For example, if your bid amount is $2 CPM and you get 10k impressions an hour, it's going to be harder on your bottom line to pay $4 for 12k per hour...

Thirdly, use every trick in the book to get maximum return from each and every visitor - back button and popunder scripts male ALL the difference to ROI - test the back button on bouncing direct to the offer/ scarcity popup on same lander / or to a new lander and offer entirely.

Lastly, make sure to compress all your image files to the smallest they will go - https://kraken.io/ is a great tool that will crunch those image files sizes and give you faster load times.

There are a lot more tips that have not been mentioned so far, will come back and add some more when I get chance. To be honest I could happily talk about POPs all day, a real favourite traffic source that is strangely addictive!


03-10-2016 08:40 PM #14 jroes57 (Member)

When you say to test different bids level, do you do this before or after optimization of placements?


03-10-2016 08:50 PM #15 simon_89 (Member)

Quote Originally Posted by jroes57 View Post
When you say to test different bids level, do you do this before or after optimization of placements?
In the beginning of your campaign, I'd suggest bidding low just to see if your offer works. However, don't bid too low to the point that you're ad is the last to be shown. As Amy mentioned about bidding in one of her posts,

0.0084 - 18000 impressions - 1st bidder
0.0074 - 18000 impressions - 5th bidder
0.0064 - 5000 impressions - 13th bidder

In this case I would bid between 0.0084 and 0.0074 which would ultimately be 0.0079 to get a fair amount of impressions.

Funnel Optimization(my opinion):
1st work on getting the best offer - 2nd work on getting the best lander with offer - 3rd then you can start playing with bids - 4th once you find the best bidding point then you should start cutting placements.


03-10-2016 09:07 PM #16 jroes57 (Member)

Should this all be tested at once and just judge the impressions per hour? and I assume it involves making seperate campaigns to test it all


03-10-2016 09:11 PM #17 simon_89 (Member)

Quote Originally Posted by jroes57 View Post
Should this all be tested at once and just judge the impressions per hour? and I assume it involves making seperate campaigns to test it all
Can you elaborate on this? Do you mean testing the staggered bids? If so, go ahead if you have the budget to do that. But, for me I like to test a campaign to see whether or not it even converts. So it's sensible for me to test at a low bidding point first.


03-10-2016 09:39 PM #18 randomman (Member)

Great thread. Subscribed.

Do you add scripts AFTER you find out if the offer converts?

To me the most important variable of the funnel is seeing whether the offer converts before you do all the testing on landers, bids, etc.


03-10-2016 11:04 PM #19 vortex (Senior Moderator)

Quote Originally Posted by dazed1 View Post
The first was buying my traffic in bulk without specifying wifi or carrier only. It's far better to let your tracker do the optimisation and send certain portions of traffic to different offers which match the visitor type. As Seb pointed out earlier, carrier traffic is where the action is at - use your highest paying and strongest converting pin-submit offer here and make the most of it! (Look for one-click flows) It's worth noting that if you can't find a good fit for wifi traffic then consider passing it on to a network such as Monetizer or YTZ.com
Great points! Just want to add that not every traffic network has carrier targeting either. And even on popads where you can target by carrier, you'll sometimes STILL get more wifi traffic than carrier even if you specify "cellular" or "cellular/carrier" as both "Connection Type" and "Connection Speed". In those cases you have no choice but set up tracker rules/flows to optimize wifi and carrier separately.

For that wifi traffic, Monetizer and YTZ as you've mentioned are great. There are also traffic networks that will buy that traffic - such as plugrush for adult pop. And then once in while when you come across a really good offer that converts well on wifi - CHA-CHING!


Thirdly, use every trick in the book to get maximum return from each and every visitor - back button and popunder scripts male ALL the difference to ROI - test the back button on bouncing direct to the offer/ scarcity popup on same lander / or to a new lander and offer entirely.
Gold! Added to list!


There are a lot more tips that have not been mentioned so far, will come back and add some more when I get chance. To be honest I could happily talk about POPs all day, a real favourite traffic source that is strangely addictive!
Please do - thanks for sharing such valuable tips. You guys have turned this into an amazing thread!



Amy


03-11-2016 01:36 AM #20 xxf8xx (Member)

Something that got the best of me as a noob is being impatient and turning one camp into something else. Eg. I would have a low bid camp and sometimes I would mess with the bid all day just to see what happens because I am too impatient to wait for the TS to approve multiple camps. From now on I will be trying to avoid this because it basically means I bought a lot of useless data.


03-11-2016 01:53 AM #21 vortex (Senior Moderator)

Quote Originally Posted by xxf8xx View Post
Something that got the best of me as a noob is being impatient and turning one camp into something else. Eg. I would have a low bid camp and sometimes I would mess with the bid all day just to see what happens because I am too impatient to wait for the TS to approve multiple camps. From now on I will be trying to avoid this because it basically means I bought a lot of useless data.
Actually this is not a bad approach. If you're testing landing pages and offers, the impact of the changing bid on traffic quality etc. applies equally to the entire funnel. So your test data should still be good.

If your goal is to map out a bid vs. volume chart, or to see which bid will give you the highest ROI, this approach will give you better results if your geo has a lot of traffic and you're getting a lot of conversions. Say you're switching the bid every few minutes, and finish your sampling within the hour - the typical time-of-day fluctuations in traffic volumes and conversion rates will probably not be huge over that hour.

However, if you're adjusting the bid randomly throughout the day and comparing the respective traffic volume or ROI that way - your results will be affected by time-of-day differences, so you wouldn't be comparing apples to apples.

This is why I prefer to just set up a few test camps at staggered bids to start - results are more accurate, plus it's less work for me - basically set a low daily budget on each and review results when I have time.



Amy


03-11-2016 02:22 AM #22 webdev (Member)

Although this is a good beginner cheat-sheet, following this guide ONLY will not put you into the green. Pop traffic is not very stable, is low quality (many adv won't allow this, or they decrease payout significantly), many traffic sources are running there own campaigns, many offer/lander combo are transparent. But I am sure following this guide you will definitely build a strong base find pockets of money!! But to make stable xxx - x,xxx/day your going to need a EDGE!


03-11-2016 02:24 AM #23 vortex (Senior Moderator)

Quote Originally Posted by randomman View Post
Great thread. Subscribed.

Do you add scripts AFTER you find out if the offer converts?

To me the most important variable of the funnel is seeing whether the offer converts before you do all the testing on landers, bids, etc.
I guess that would depend on your style. I like to start with an offer type and traffic type, put a bunch of offers and landers into an initial test to see how close I am to profits, and decide whether or not to continue.

So I test landers from the start. I want to evaluate how easy it would be to run that vertical for that traffic type. Testing multiple offers and landers will give me a better picture. If I just tested a single offer using a single lander, and if they don't perform, that doesn't tell me anything - may mean the vertical on the traffic type is hard to make profitable, but could also mean the offer or lander just happens to be bad. I wouldn't know.

As for bids: I've had initial tests that did horrible ROI at low and average bids, but were profitable at higher bids. Without testing bids from the beginning, I would have dismissed the camp as being unprofitable and moved on.

Thank you for your question!



Amy


03-11-2016 05:10 AM #24 ayuluv (Member)

thanks Amy for your sharing

on the side note, how do I actually input tracking token so that I will be able to track in Voluum?


03-11-2016 05:38 AM #25 ayuluv (Member)

Quote Originally Posted by sebastian_r View Post
Nice write up!

Here's my take:

Speed, Speed, Speed.

Invest into the right infrastructure before running pops. A shared 10$ server won't fit the bill if 50k impressions per hour are hitting. I have an DVS 12GB from Beyond Hosting and use Cloudflare for DNS. The combo was able to handle x,xxx,xxx impressions per day.

Optimize Lander speed. Should not be bigger than 100-300kb. The quicker the page loads, the higher your chances to convert. The difference between .1 and .15 in CV can already make or break an pop camp.

Non aggressive won't fly. I've spent $xxx,xxx on pops and did not have an single non aggressive camp working.

The pop traffic sources are all stealing from you (I consider 50% bot traffic stealing). Cutting like a madmen is the name of the game. 250 impressions without click, 1000 impressions without conversion for low payouts (stretched throughout the day). With high payout pins it can be a bit more.

If you hit something hard, most networks or their employees will copy your camp. Cloak.

Antivir, iphone, whatsapp, and utility apps are doing volume. The rest is a side show.
Thanks mate for your input
How do I actually cloak if I need to?
Thanks!


03-11-2016 07:50 AM #26 mling1985 (AMC Alumnus)

Thanks for posting this cheat sheet and all your awesome posts Amy, I have learnt a great deal from them.

Got a question for you. You have always stressed not to cut too many placements without improving the funnel first.
But do you have a general rule of thumb cutting them starting a RON campaign when there are lots of placements that are low quality?

Cheers


03-11-2016 08:19 AM #27 sebastian_r (Member)

If you're running on a low budget, you can cut early (worse than -20/-30% ROI), improve your funnel on the go, and later retest the bad placements.


03-11-2016 08:37 AM #28 lanikai87 (Member)

This is a nice cheat sheet. Thanks Amy.

I love pop traffic


03-11-2016 09:10 PM #29 vortex (Senior Moderator)

Quote Originally Posted by ayuluv View Post
thanks Amy for your sharing

on the side note, how do I actually input tracking token so that I will be able to track in Voluum?
When you're adding the traffic source to Voluum, click on "Advanced" (with the little gear icon) in the settings for the traffic source, and you'll see a list of empty fields where you can put in those tokens.

"Parameter" and "Name" aren't important, but it would help you to tell them apart if you gave them meaningful names. The "Placeholder" is where you'd put the tokens from the traffic source. Lastly, don't forget to checkmark the box at the end so voluum would know to track the variable.





Amy


03-11-2016 09:19 PM #30 vortex (Senior Moderator)

Thanks for posting this cheat sheet and all your awesome posts Amy, I have learnt a great deal from them.
You're welcome! I'm picking up tips from all the feedback as well. This sort of collaboration is what makes STM such a great community.


Got a question for you. You have always stressed not to cut too many placements without improving the funnel first.
But do you have a general rule of thumb cutting them starting a RON campaign when there are lots of placements that are low quality?
I agree with what sebastian said:

If you're running on a low budget, you can cut early (worse than -20/-30% ROI), improve your funnel on the go, and later retest the bad placements.

There's more than one approach to doing any one thing, and it would really depend on the particular situation. The main consideration is how much traffic you're able to get at the moment. Because I work mostly with new affiliates, and because most of them start with smaller geos and often lower bids, they can't afford to cut placements early, or else they wouldn't have traffic left to test offers and landers.

Another consideration: If a traffic source has a lot of bot traffic, of course you'd want to cut more aggressively from the start.

This may be overkill, but I've gone into this topic in great detail in this post:

http://stmforum.com/forum/showthread...l=1#post248980


I love pop traffic
I do as well!


03-12-2016 08:02 AM #31 mling1985 (AMC Alumnus)

Quote Originally Posted by vortex View Post
There's more than one approach to doing any one thing, and it would really depend on the particular situation. The main consideration is how much traffic you're able to get at the moment. Because I work mostly with new affiliates, and because most of them start with smaller geos and often lower bids, they can't afford to cut placements early, or else they wouldn't have traffic left to test offers and landers.

Another consideration: If a traffic source has a lot of bot traffic, of course you'd want to cut more aggressively from the start.

This may be overkill, but I've gone into this topic in great detail in this post:

http://stmforum.com/forum/showthread...l=1#post248980
Thank you Amy, that post you linked have really cleared things up for me, I dont know how I have missed it before

Got 1 more question, assuming budget isn't that big of a concern, how many impressions daily would be a good amount to test our funnels?


03-15-2016 07:11 AM #32 vortex (Senior Moderator)

Quote Originally Posted by mling1985 View Post
Thank you Amy, that post you linked have really cleared things up for me, I dont know how I have missed it before

Got 1 more question, assuming budget isn't that big of a concern, how many impressions daily would be a good amount to test our funnels?

Very hard to say - without knowing the vertical, geo, traffic source, your bid etc.

Basically the more traffic you get, the sooner things reach statistical significance and the faster you can cut them to optimize your funnel. So as long as the traffic converts well enough that you're not losing an arm/leg (e.g. by bidding so high that you're paying unreal amounts just to get more impressions), I would say the more impressions the better!

If you're getting a lot of traffic, though, you'll want to check your data for statistical significance more often, in order to cut stuff when they're ready to be cut. Otherwise you'll "overrun" your split-testing and waste money unnecessarily.



Amy


03-15-2016 01:32 PM #33 mling1985 (AMC Alumnus)

Quote Originally Posted by vortex View Post
Very hard to say - without knowing the vertical, geo, traffic source, your bid etc.

Basically the more traffic you get, the sooner things reach statistical significance and the faster you can cut them to optimize your funnel. So as long as the traffic converts well enough that you're not losing an arm/leg (e.g. by bidding so high that you're paying unreal amounts just to get more impressions), I would say the more impressions the better!

If you're getting a lot of traffic, though, you'll want to check your data for statistical significance more often, in order to cut stuff when they're ready to be cut. Otherwise you'll "overrun" your split-testing and waste money unnecessarily.



Amy
Thanks again Amy , I am running sweeps SOI offers in tier 1/ tier 2 countries on adcash, bidding just above their average so around, 5 CPM and generally getting around 30,000 impressions a day after blacklisting some of the worse sites, does that sound like a healthy amount of traffic to test my funnels?


03-15-2016 03:25 PM #34 cbrughmans (Member)

Great tips Amy, thanks a lot for that.
One more very important thing is to make sure you have a cookie duration of +30 days with your advertiser. A lot of conversions will come in on the long tail when you run pops.

Verticals that work well on pop traffic are:
- gambling and sports betting
- desktop downloads
- big brand names (the sun, nowtv, etc)
- video on demand (free trial offers)
- sweeps


03-16-2016 04:51 AM #35 sushiparlour (Member)

Quote Originally Posted by cbrughmans View Post
Great tips Amy, thanks a lot for that.
One more very important thing is to make sure you have a cookie duration of +30 days with your advertiser. A lot of conversions will come in on the long tail when you run pops.

Verticals that work well on pop traffic are:
- gambling and sports betting
- desktop downloads
- big brand names (the sun, nowtv, etc)
- video on demand (free trial offers)
- sweeps
Interesting point cbrughmans, any chance you know of any article or posts in setting this up as I wasn't aware conversions were based on cookies (thought they had to convert during the same session or it becomes void)


03-16-2016 12:24 PM #36 cbrughmans (Member)

Conversions are attributed by most advertisers on a 'last-click-counts' basis. So if you are the ad that the user clicked on for the last time before converting, then the commission (CPI/CPL/CPA) is yours. If the user first clicked on your ad, then on someone's else's ad (for the same advertiser), the conversion gets attributed to that person and you will get nothing.

Lots of advertisers - and especially big brand names - have massive amounts of organic traffic. If your cookie is still on the browser and as long as its still active, you will get your commission paid out for all that organic traffic as well according to the last-click-counts model. Because of this reason lots of (big brand) advertisers don't allow pop traffic as they see it as cookie-dropping/stealing their organic traffic.


03-16-2016 02:46 PM #37 Wiget_Team (Member)

Great thread guys!

From our perspective (as one of the biggest mainstream popunder platforms), I can say we rarely see guys go from $10 a day to $1 000 a day, while on the other hand we oftentimes see the $500 a day go to $1k - $5k+ per day. I really think the testing phase is super important when it comes to pops as a channel. You need to try out a few GEO's, offers, angles etc., and it's hard to do that on a super tight budget. Usually it pays to start off a bit broader and then scale on the sweetspots.

If you want to give the mainstream pops a go, it's a great time now at Wiget since we offer all STM members 25% bonus on top of their first deposit. You can also hit me up on PM if you need any assistance, or have any other questions on popunder traffic.


03-16-2016 03:16 PM #38 cbrughmans (Member)

Agree with the above point. Normally I always recommend a 100$ per campaign budget for testing purposes but Pops is such a big channel that you would need to test at least 300-500$ to get any valuable insights and enough data to determine your possilibities to scale up in a profitable way.


03-16-2016 04:34 PM #39 caurmen (Administrator)

Pops is such a big channel that you would need to test at least 300-500$
I'd disagree with that as a sweeping statement.

Required budget to test is almost exclusively a function of number of variables in the traffic source (mobile has more than desktop) and payout of the offer, as well as just how many iterations of testing you're willing to go through. Run test -> reach statistical significance -> repeat. The amount of budget you need to reach statistical significance is usually a function of the payout, as your eventual goal is positive ROI. The potential maximum volume of a channel really isn't relevant except as a potential upside.

The only exception to that rule (and this may be what Cristoph means here) would be if you're expecting a traffic source to have a lot of crap traffic and a few gems, and you're running an RON test to filter out the gems. Then you'll need to spend significantly more because you're effectively running dozens or hundreds of tests, not just one.

(Indeed, although I'm getting slightly off-topic into advanced optimisation here, in many ways pops are easier to test than display, because you have less stages to test: where display goes banner -> lander -> offer, pops often just go lander -> offer. That opens up possibilities for a variety of data-hungry optimisation strategies.)

Vortex's guides to cutting landers and placements are excellent - http://stmforum.com/forum/showthread...Banners-Part-2 - and we've also got detailed coverage of optimisation strategies over at the Getting Started Guide - http://stmforum.com/forum/forumdispl...g-2016-Edition


03-16-2016 07:39 PM #40 bradh_ ()

Thanks for the awesome guide Amy! I'm still a newbie, but following just a few of these tips got me to my first ever breakeven campaign. Only up from here!


03-17-2016 07:35 PM #41 brodycurtis ()

Hey Amy,

Thank you for this great information! It was a great read


03-18-2016 03:55 PM #42 vortex (Senior Moderator)

Quote Originally Posted by mling1985 View Post
Thanks again Amy , I am running sweeps SOI offers in tier 1/ tier 2 countries on adcash, bidding just above their average so around, 5 CPM and generally getting around 30,000 impressions a day after blacklisting some of the worse sites, does that sound like a healthy amount of traffic to test my funnels?
Oh yeah 30k/day for SOI sweeps offers should allow you to test offers and landers very quickly! As I've said before, try to check several times a day for statistical significance to make sure you cut stuff as they're ready for cutting. Also, with that amount of traffic, you could even just run traffic during peak hours of every day when conversion rates are highest, just to save money on testing. In many geos these will be the after-work hours - often from 5pm onwards.

If test budget isn't a concern then by all means run 24/7 - the faster you gather data the faster you'll hit green and the sooner you can scale!


Amy


03-23-2016 03:07 PM #43 brodycurtis ()

Quote Originally Posted by vortex View Post
Here's a more newbie-friendly version of the above, with certain points explained in greater detail.


1. Popular mainstream pop networks: Zeropark, PopAds, PopCash, Adcash, PropellerAds, WigetMedia, Gunggo, Mediahub, SelfAdvertiser.

Networks with adult pop traffic: Zeropark, PopAds, Exoclick, JuicyAds, Ero-Advertising, Traffic Junky, Traffic Force, Traffic Factory, AdamoAds.



2. Run broad-appeal offers. Since pop traffic is non-targeted for the most part (i.e. doesn't allow targeting by user demographic), make sure your offer will appeal to the general populace.



3. Don't dismiss low-payout offers. Pop traffic can be purchased at relatively cheap rates (compared to other traffic types). Even offers with $0.1x payouts can make you good money, if the target geo has cheap traffic prices that convert well.



4. Check how aggressive you can get. Most of your competitors will be running aggressive, so you need to do the same if possible.

Check with your affiliate network on how aggressive you can get when pushing the offer. Also check your traffic source's creative requirements to confirm your landers are in compliance.



5. Test low-competition geos first. Some geos are easier to make profitable than others. If you want to see green faster, try geos that have lower average bids. You can find this information in the traffic source's inventory report or traffic estimator. Alternatively you can ask their support.

Also make sure the traffic source has enough traffic in that geo, for your targeting options, to be worth your time.

As well, some sources will also tell you how many advertisers are running in each geo (e.g. PopAds), which is another way to gauge competition levels.

Where to look? Asia, Africa, & LATAM America



6. Test popular landers that look different, to establish benchmark stats. When testing a new vertical on pop, with have no proven landers, test 5-10 of the most popular landers to get an idea on what range of CTRs and CRs you can expect. Of course this will depend on the offer as well, so testing multiple offers from the start would be wise too.

How to find popular landers? Do some manual spying (thank you Simon!), or use a spy tool like Adplexity.



7. Make landers attention-grabbing. Because pops are a form of "interruption marketing", you need to use every trick in the book to turn the visitor's focus from what they were doing originally, to your lander, and then hopefully to your offer.

If the offer and the traffic source allow it, use aggressive landers, and test the common CTR/CR-boosting tricks: entry/exit alerts, backbutton redirect, audio, vibrate, etc.

Optimize page load speed by using a CDN and by optimizing the coding.



8. Test multiple bids from the start. Some traffic sources will provide an "average" or "recommended" bid, but bidding at the average will not always give you the best value for your money.

Make copies of the new campaign so you have 3-5, and set a different bid for each ranging from low to average to high, set a low budget for each (10-20x payout should be enough). Run traffic, see which bid gives the best ROI, and use that camp to test your landers and offers.

Testing different bids will allow you to a)gauge promise of offers and landers, and b)save you money by testing at a bid that give good ROI.

There's no set method for selecting bids, but see here for a sample approach for popads.



9. Don't be afraid to test high bids. On many pop sources, higher bids can get you traffic that converts better. I've had initial test camps that performed vastly different at high and low bids, where the camp was profitable at the high bid but in great loss at a lower bid. If I hadn't tested the high bid, I would have written the campaign off.

Important: Remember to set a low budget as a high bid doesn't always translate into high traffic quality!



10. Get rid of bad placements. Pop sources typically have a ton of publishers/placements, and they don't all convert the same (some don't even cost the same). On certain pop sources where bot traffic is rampant, you may need to pause quite a few placements to see green. See here for a tool to help identify bad placements.

Before cutting placements en masse though, please see the next point...



11. Improve your funnel first, cut second. When evaluating a placement, remember that even the best placement won't do well with a bad offer and lander. The same applies to other targeting, e.g. OSs, devices, carriers, browsers, time of day. So split-test offers and landers until you find a combination that can make a good chunk of your traffic profitable, and THEN cut anything that isn't likely to give you the ROI you want.



12. Set a frequency cap, at least in the beginning. More often than not, turning off frequency will bring down conversion rate. Before your camp reaches profitability, it would be wise to set a frequency cap to e.g. 1 in 24 hours.

Once the camp is profitable and doing good ROI, test higher frequencies, or even turn off frequency altogether. The conversion rate may drop, but if your overall profit amount is higher than before, then that's all you'd care.



13. Test other available ad formats. For example, zeropark has both pop and redirect traffic, and popads have popunders, popups, tabunders and tabups. If you have an offer & lander that convert, try the other formats on the same traffic source.



14. Use traffic network tracking tokens to help you track the performance of traffic segments you can target at the source. This will let you optimize for traffic segments that are giving you the most traffic and the highest ROI. You can find tracking tokens for many networks here: TrackingTokens.com.






Amy
I would add TrafficVance which is now called Propel Media ( https://www.propelmedia.com/advertisers/) to the list of top mainstream networks.


04-11-2016 11:50 PM #44 exclusif (Member)

Great thread. "Aggressive" is hinted at quite a bit, do you include false promises in that when it comes to sweeps? As in "You've won" instead of "Here's your chance to win"? If so, how does that affect lead quality?

Also, anyone got some affiliate networks with a lot of sweep offers for 2-3 tier countries? Bit disappointed in Clickdealer and MundoMedia in that specific regard.


04-14-2016 08:29 PM #45 vortex (Senior Moderator)

Quote Originally Posted by exclusif View Post
Great thread. "Aggressive" is hinted at quite a bit, do you include false promises in that when it comes to sweeps? As in "You've won" instead of "Here's your chance to win"? If so, how does that affect lead quality?

Also, anyone got some affiliate networks with a lot of sweep offers for 2-3 tier countries? Bit disappointed in Clickdealer and MundoMedia in that specific regard.
Yes - aggressive usually means misleading, or making false promises as you put it.

With sweeps, there are offers that allow aggressive, and then there are ones that don't. The ones that do allow aggressive - the advertiser would be more prepared to accept lower lead quality usually. Actually for sweeps I may need to revise my recommendation to only run offers that allow aggressive, because saying "you stand to win" instead of "you've won" can perform approximately the same in terms of conversion rates. I've seen it happen more than once already. So if lead quality is an issue and/or you're wanting to promote a sweeps offer that doesn't allow aggressive, by all means test compliant text and see.

I've never run sweeps heavily and the last time I ran them was a long time ago. You can try adsimilis and oasisads. As far as I remember, there just aren't that many sweeps offers for tier 3+ geos. Maybe do a search on offer aggregators like odigger, or take a look at makemassive, to find out where they are.



Amy


07-14-2016 12:35 AM #46 Zartan (Member)

I have a very aggressive lander for adult and have an almost 11% CTR but -100% ROI.

Wondering if it is the offer or if my lander is incongruent with the offer's opt-in page?


07-14-2016 01:08 AM #47 affpayinggao (Veteran Member)

Excellent info! Thanks for sharing Amy!


07-14-2016 01:38 PM #48 matuloo (Legendary Moderator)

Quote Originally Posted by Zartan View Post
I have a very aggressive lander for adult and have an almost 11% CTR but -100% ROI.

Wondering if it is the offer or if my lander is incongruent with the offer's opt-in page?
There is a disconnect somewhere, or the offer is not good at all - can be both really.


07-15-2016 01:55 AM #49 vortex (Senior Moderator)

Quote Originally Posted by Zartan View Post
I have a very aggressive lander for adult and have an almost 11% CTR but -100% ROI.

Wondering if it is the offer or if my lander is incongruent with the offer's opt-in page?
As matuloo said.

Also, just because a lander is aggressive doesn't automatically mean it will convert better.

Making the transition from lander to offer page more congruent will help. But I would suggest to do some spying in the wild first to see what types of landers are being used to promote your type of offer, and testing those first to find out what type of lander converts best for your offer, and go from there. Finch provides a step-by-step on how to spy on your competition in this excellent guide:

http://stmforum.com/forum/showthread...Dating-in-2015


CR and not CTR would be the way more important metric when comparing landers. If a lander doesn't convert it doesn't matter how high its CTR is.



Amy


04-02-2017 12:02 PM #50 ActiveRevenue (Member)

Wow! amazing post Vortex. Thanks a lot. I also would like to add ActiveRevenue as a mainstream traffic source. Still young (Previous was AdventureFeeds) but getting to the highest levels very fast.


04-05-2017 05:55 AM #51 callmebix (Member)

this is my second to read your all thread,every time i will learn new knowledge.
I have a problem now,I use the traffic source is common, such as zeropark, popads.
I think these traffic sources are too competitive and want to profit by looking for new traffic sources.
Do you think the idea is right?


04-05-2017 10:55 PM #52 vortex (Senior Moderator)

Quote Originally Posted by callmebix View Post
this is my second to read your all thread,every time i will learn new knowledge.
I have a problem now,I use the traffic source is common, such as zeropark, popads.
I think these traffic sources are too competitive and want to profit by looking for new traffic sources.
Do you think the idea is right?
Seeking out less-saturated traffic sources would definitely be a good idea! Here's a rough outline on how to approach that:

-Compile a list of pop sources - there are a few floating around this forum, and you can also google.

-Craft an email to ask each source for 1)list of highest-volume geos along with approx. daily/monthly traffic volumes, 2)minimum initial deposit amount. Some networks have a contact form you can use. Others you'd need to email.

-Pick sources that have enough volume, and sign up, make your initial minimum deposit, and test a currently-profitable camp on them. If the minimum deposit is high, talk it down or leave it for testing later. NEVER make a deposit without getting a reply back from a traffic source! Some are completely defunct and you may not even get your money back. Basically when you venture outside of the most well-known and reputable networks, you risk losing your deposit, so try to avoid the ones that require large minimum deposits until you're at a point where you can afford to lose it.

Hope that helps!



Amy


01-19-2018 08:14 PM #53 noobie2018 (Member)

What is meant by the term aggressive. for example...use aggressive landers?


01-19-2018 08:22 PM #54 vortex (Senior Moderator)

Quote Originally Posted by noobie2018 View Post
What is meant by the term aggressive. for example...use aggressive landers?
That's correct! Misleading angles, or even using tactics like iframes and such - tactics such as the ones listed here:

https://stmforum.com/forum/showthrea...ark-arts-of-am



Amy


Home > Paid Traffic Sources > POP / PPV / Redirect