Mobile pops + sweepstakes first steps follow along (5)
Hello STM!
I've written a short intro here about our team and promised a follow along..So I'll cut right to the chase 
First AM campaign ever.
Target & Budget: Initially we wanted to spend $50 to figure the basics out with sources/landers/tracking/offers and building a funnel, no split testing yet on landers or offers. We ended up spending $120 as the campaign did well.
Traffic source: Popads, won out for first campaign vs. Propeller and Popcash due to good targeting options and low min. top up $
Type: Mobile pop
Country: Hungary / we are Hungarians and figured we could fix common misspellings and translation errors on landers we find + provide very fast load times through own HUN server.
Offer: Checked 4 Mobidea Iphone 7 Sweepstake offers and tested the user flow. Ended up picking one that we thought had the best lander. $8.73 payout.
Lander: Ripped one that seemed to be popular and doing well (spin the wheel), modified it, made some changes to push down the load time, corrected spelling errors and wording inconsistencies.
Tracker:
We started with 3G traffic only, Android (very old / latest versions excluded), 1/12 freq. cap, no blacklists.
First results - out of 3 Carriers one (A) did very well in the first five days (142% ROI), two other (B&C) wasn't that much (-31% and -32%). We introduced a new offer to the other two carriers and tried splitting between the two, results improved on both offers eventually so we let that run. I've also tried to increase the bids and letting worse quality sites in for the carrier A so we can scale, but interestingly traffic started to go down (WTF??
)
Second phase - We carried on buying the now lowered amount daily on carrier A and B&C started to show some positive ROI. I've weeded out some device types and time slots that didn't bring any conversions. We introduced a new campaign buying Wifi traffic on a Friday which ran through the weekend and on Monday with exceptional results: $16 spent, $46 revenue. Carrier traffic started to slow down on conversions though. Wifi went dead in practically one day. Last two days produced a $7 loss overall so we stopped the whole campaign this Wednesday.
Overall result:
What we learned and did well? We managed to figure out putting a good traffic campaign together in Popads, adding campaigns, landers and split tests to
What we didn't learn or do well? We didn't set up passing of information like bids, siteid etc. through tokens from Popads to Voluum. We want to use these variables better next time to exclude sites & to track costs automatically (see Q's). As a result we didn't build a blacklist at all. We didn't use and track site categories either. I have to add due to low volume and traffic scaling issues this didn't seem valid anyways and it would have further lowered our numbers.
Our next steps & targets:
We plan to attack this in 3 new ways:
- Do the very same campaign but with different traffic sources - Popcash & Propellerads
- Do the same (Popads - Hun traffic - Same offer) but with a different landers to see if we can squeeze more out of it with different angles
- Do the very same campaign with Popads traffic but in different countries (prelander translated, proper hosting, corresponding Mobidea iphone sweeps etc)
Questions:
- If a campaign seem to run its course in 2 weeks is it possible to restart it with the same variables let's say in 2 months? Do you guys "recycle" exact campaigns this way? Maybe with a different offer but same funnel.
- What do you think about our next step ideas? Any other ways we can or could have built on this success and scaled this to make more money?
- Any other recommendations for traffic besides the 3 mentioned in the thread for this type of campaign?
Thanks for reading & please post if you have any recommendations
Yes it makes a lot of sense to retest old campaigns.
Whenever you see new offers for old campaigns you should give them a new run.
Just let the campaigns stand on the traffic source and
Easiest way to scale a campaign is getting more traffic from the original traffic source. Break out profitable targeting options into separate campaigns with higher bid to get more traffic, bid into the dark to discover new placements, 2. and 3. campaign or 2. and 3. account.
Then scaling to new pop sources, here you will have to test a lot to find some that work. Test 10 find 2-3 winners. Some that you haven't mentioned: Zeropark, Adcash, Widget, Mediahub, Selfadvertiser, TONIC
Then it gets already difficult, scaling to a new GEO. This is all about finding another great offer. Best way to tackle this task is to start with the offer then build your campaign around it. Talk to your AMs and ask for well performing offers in your vertical without predefined GEOs. Then find traffic and landing pages for it.
"First AM campaign ever."
wow that's cool .
i am following this to see what's going to happen in the next days .
Sebastian, appreciate the detailed response and the great tips. We will keep all this in mind especially the when to reactive tip & also the other pop sources + success at other geo's depending on the offers.
Here is a quick update:
- We tried to replicate our success with Popads + HU and try Popcash & Propellerads as sources. Popcash yielded a 7 USD profit (+38% ROI), Propellerads 9 USD loss (-19% ROI) after around 15K hits each. What we learned is we probably should have introduced these earlier in the process. Stopped.
- We are currently running campaigns in Popads in 1 other country and 1 coming up with an adapted lander (English speaking both), will follow up with the results in a couple of days.
- We are working on making our proprietary prelanders, the first one in quiz style to test with the original HUN offer + source. A question here to anyone out there: can you "reboost" a slowly dying campaign by introducing a new landing page if the source/offer is the same? Any experience with that? Two ways we are thinking is either doing a prelander which is similar in style but different content (eg. a quiz instead of a spinning wheel) so the user who saw/converted with the previous might think it's a new chance for them (retargeting could work as well), the other is to build a 100% different lander.
Thanks for reading and good luck hunting down those conversions 
You can try to revitalize the campaign with a new lander, however the offer has the bigger leverage.
When the campaign performance suddenly drops, it's in most of the cases the offer.
You should first try to work the variable that causes the performance drop, then try to work the other variables.
-> Test more offers and find a better performing one
Home > Mobile > Follow-along Campaigns