PIN submits & lead quality - prepay leads (7)

Just got kicked off an offer for sending prepay leads. It was a PIN submit in a mid-sized GEO, and my main chunk of traffic was WiFi POPS - which was converting nicely.

AM gave me a heads up and I'm not sure what to do with this or how to interpret this feedback:

AM: Advertiser hates your traffic man.
AM: Wants you to pause.
- kk
AM: What are you sending?
- popcash, most convs from one page
AM: Thats 2 advertisers now
- pausing it all
- what stuff can i do to improve the lead quality? you have any suggestions?
AM: Prob improve the source to be honest
AM: Their sending you prepay shit
- that excludes most pop sources
- all i have worked with so far is popads popcash propeller adcash selfadvertiser and so on
AM: actually pop does work well - but they need to know how to white/black list the placments.
It's usually due to too much bot traffic
- do the bots provide fake leads sometimes?
- are there some parameters i could pass to your platform so help the advertiser optimize my traffic?
AM: Best Bet is to go through STM and see
AM: They would know more than I or advertiser about placements and blacklisting.

I'm not sure what this means, send only carrier traffic and exclude all the WiFi traffic? My main source of profit on PINs so far has been WiFi, carrier volume has never been significant : / I got this feedback after getting up to a total of 94 leads over a bit less than a week - CVR at network side is 1.6%.

ask for clickid of bad leads, check from which widgets/placements they come, block the widget/placement. if they all come from one carrier, send the specific carrier traffic somewhere else.

Wifi usually doesnt work well for PINs at all, so you've hit something unusual which might have been the reason for poor traffic quality. Sebastians suggestion was a good one, you need to identify where exactly did those "bad" conversions come from. Hopefully they can give you that info, which isnt that easy sometimes.

Wifi works perfect for pins, but you need to make sure you run proper devices, which have sim cards. Eg -exclude anything resembling tablet that dont have a sim card - if we are talking about a properly optimized wifi offer which has mo-click. If we are talking about an improperly optimized offer (merchant's fault or legal restrictions) then it runs on number submit. If the offer is number submit (not mo-click) you have a slightly better chance of converting on a tablet, but its still a bit fucked up. Since surfer surfing on sim-less tablet inputs his mobile phone, and he usually needs to confirm the sub with a reply sms. Unless he has the phone with him, his conversion drops.. a lot.

Now with regards to the op's problem and am's advice, which is actually pretty sound, even tho the am does not understand or didnt want to say the underlying fundamentals for pops sucking (on wifi)

The problem you got into is purely becuase of demographic.

Majority of pop by volume runs on livestreaming, torrent and similar crappy sites. These sites are visited primarily by younger demographic. In some countries younger demographic is primarily on prepaid. And then you get an exponential effect of crap on crap on crap, resulting in crappy backend performance. Now if you filter out all of the placements hit by youngsters on prepaid and get placements such as for example a popunder on a iamanoldfartwithdiabetes.com or seniorserectionproblem.com or cmn.com (a typo of cnn with some filler content) ** all domains are fictional, where the youngester on prepaids are not so prevalent you will be well off.

Now if you take pops and carrier traffic and dump them to pin offers (non dcb - so no carrier billing), it prolly does indeed work better, becuase your typical prepaid user will not have a data plan in place and thus will not be on carrier.

To identify and eliminate placements that have high percentages of bot traffic:

http://stmforum.com/forum/showthread...n-Any-Campaign

The approach I'd use, would be to start a "bot test" camp, set a really low daily budget (the lowest the traffic source would allow), wait for that budget to run out, check stats and cut all placements that have reached 60 impressions no matter what % bot traffic each has, then move your budget up by another couple dollars and run again, cut more placements, and repeat. You'll find that you won't need to repeat too many times, or spend too much money, in order to identify the biggest and worst placements. (You don't need to go through ALL placements - just the bigger ones should be enough. Otherwise it would take forever and you'd get diminishing returns on time and money spent.)

Note: I've found that some of the best-converting placements have high percentages of bot traffic, so I normally wouldn't recommend cutting placements that have say 20%+ bot traffic. But in your case it's not conversions that aren't working for you - it's quality.

You can try explaining your plan to your AM and see if you could take another stab at the offer again. If the advertiser has "banned" your aff ID, ask your AM if you could register for a separate account on the affiliate network and try running better-quality traffic and see what will happen.

Or: Look to see if the same offer is available on another network and test that.

Or: Just run other offers - not every advertiser is as strict about quality. Some advertisers have higher margins because they monetize on the backend better. Some advertisers are also more savvy in pricing their offer payouts to take varied lead quality into account.



An alternative approach would be to target JUST carrier traffic, and test offers that are designed for such traffic - i.e. one-click carrier-billing offers:

http://stmforum.com/forum/showthread...esting!-Part-1


Good luck my friend! You've been making tremendous progress. Kudos for not letting this deter you.



Amy

pass back the zoneid to the advertiser and let them optimise it for you, that way you have more transparency and there's no bs. If you dont want to expose your zoneid's your gonna have to do it the hardway and figure it all out for yourself. I found that when you work closer with the advertiser they usually like you more and when the quality goes up so will your payout.

Originally Posted by johner911

Wifi works perfect for pins, but you need to make sure you run proper devices, which have sim cards. Eg -exclude anything resembling tablet that dont have a sim card - if we are talking about a properly optimized wifi offer which has mo-click. If we are talking about an improperly optimized offer (merchant's fault or legal restrictions) then it runs on number submit. If the offer is number submit (not mo-click) you have a slightly better chance of converting on a tablet, but its still a bit fucked up. Since surfer surfing on sim-less tablet inputs his mobile phone, and he usually needs to confirm the sub with a reply sms. Unless he has the phone with him, his conversion drops.. a lot.

Now with regards to the op's problem and am's advice, which is actually pretty sound, even tho the am does not understand or didnt want to say the underlying fundamentals for pops sucking (on wifi)

The problem you got into is purely becuase of demographic.

Majority of pop by volume runs on livestreaming, torrent and similar crappy sites. These sites are visited primarily by younger demographic. In some countries younger demographic is primarily on prepaid. And then you get an exponential effect of crap on crap on crap, resulting in crappy backend performance. Now if you filter out all of the placements hit by youngsters on prepaid and get placements such as for example a popunder on a iamanoldfartwithdiabetes.com or seniorserectionproblem.com or cmn.com (a typo of cnn with some filler content) ** all domains are fictional, where the youngester on prepaids are not so prevalent you will be well off.

Now if you take pops and carrier traffic and dump them to pin offers (non dcb - so no carrier billing), it prolly does indeed work better, becuase your typical prepaid user will not have a data plan in place and thus will not be on carrier.