Home > Mobile > Follow-along Campaigns

---

From Poker to mobile pops (11)

---

Hi everyone

Been a member of STM for a few months now, made a follow along previously for mobile display but decided to switch to pops after giving it a shot. I enjoy working with landing pages a lot more than banners, and it removes a variable to makes it easier to learn imo.
Anyway, I'm a 25 year old online poker pro in Bangkok trying to make it in affiliate marketing. I decided to quit poker until end of August and focus 100% on Affiliate marketing till then, and re-evaluate my situation after. I was struggling pretty hard until now, tested out 20-30 offers with 10 only conversions total :s. Was getting quite frustrated but got into a new network and the manager recommended an offer which is showing some promise. I'm also talking to "splitpersonality" on a daily basis who has his own follow-along here and learning a lot that way.


Offer : iPhone PIN submit
Traffic Source : Popads
GEO : India

Overall Stats:


By OS :


I noticed pretty early on that iOS was performing significantly better than Android, despite the lower CTR. I used this tool : https://win-vector.shinyapps.io/CampaignPlanner_v3/ to check whether it was statistically doing better and it was. It also makes sense that iOS would do better as people that already use Apple Products are more likely to want an iPhone rather than Android users (speaking as a stubborn Android user, I can confirm ).
I decided to only send traffic to iOS and Tizen.

Landers : (Drilled-Down for iOS only)

Unfortunately, none of the landers are profitable I cut the "Congrats" and "Selected Device" landers pretty early on because of the 0 and low CTR. I gradually cut the rest and I'm only running Spin, Google Rewards and WheelSpinPlay right now. The reason WheelSpinPlay has such low visits so far is because I initially wanted to run it but forgot to add it in Voluum. Since the Spin lander is doing the best and this one is similiar, I expect OK results.

Unfortunately there isn't really any green anywhere so I'm not sure whether I should spend much more time on this. It is however my first campaign with any kind of data, so I'm more focused on learning how/what to cut, the approach for doing it and potential funnel optimisation. I could try some new landers and optimise them and maybe get 50% ROI or something.

The main problem is that this offer only works with 3 cellular operators, and only running iOS even with a higher than average bid, I only get 5000 impressions per day. So my total spend per day is 20$ or so. Again, not a huge deal imo since I'm not looking to make big bucks right now, just learn and let the money come in later. I've thought about trying it on another traffic source but figured I would wait until I've got a bit more data on which landers perform best.

By Website Quality :



I know this is only a little data so we can't make too many conclusions, however I find this stuff interesting. I've been thinking a lot about which settings to choose for website quality in Popads. I'm often running at 5+ (Top 60%) of websites but I'm really unsure on what I should choose. Does anyone have any insights on this ?

----

The plan moving forward :
- I'm gonna keep getting data on this campaign and see if my new landers gets close to breakeven or not. If my landers are getting breakeven, I will try some modiications to them to see if I can get them to a positive ROI. I'm thinking speed optimisation, changing the language and just cleaning up some details or adding things such as calling out user brand/model and city etc...
- My AM gave me 3 more similiar sweeps, one in TH, one in MX and another in Brasil. Fortunately the offers are very similiar so I already took the best 4 performing landers from this campaign and getting them translated to thai. The MX offer has a somewhat high payout (3.60$) so I will wait until I gather more lander data before launching that one. Brasil payout is pretty low and I have the same offer on another network so I'm gonna get started on translations for that right away.

Questions :

- My AM told me I could run these offer "aggressive". But when I checked the offer in the network's pages, it says not allowed to use misleading creatives such as "You have won". I thought aggressive meant saying just that. Is this one of those things where the advertiser just likes turning a blind eye ? I figured I should get things clarified with the AM/Advertiser if I intend on running bigger traffic. Just wondering about this because another AM sent me an offer from the UK which states the same thing, but on Adplexity everyone is running that ad aggro stating the user has already won.

- How much of a ROI increase can I expect with funnel optimization ? So if my lander is currently breakeven, what's a reasonable ROI expectation for it after improving ? I remember reading somewhere here that people changed a picture or 3 words and it made 50%+ differences in ROI which seems quite surprising to me, yet very exciting too

- I'm currently only running popups and no popunders. I hadn't even really considered running them. Should I also test popunders ? If yes do I just run both at the same time or make 2 separate campaigns ? Can Voluum identify between a popup and a popunder ?

---

Thanks a lot for reading ! Criticism/comments highly encouraged and if anyone is in BKK and wants to grab a drink or talk about AM or whatever feel free to send me a PM

---

Today's Stats :



Not the best. Paused the India offer after it had spent 8x the payout with 3 conversions. Spent some time optimising the LP (brand+model callout, got more agressive with the wording of the page, changed the comments section a bit (changed UK names to indian ones) and made some speed optimisations). Will split-test the LP now with the old one. Also ripped another Landing Page that I will try.

I launched the Thai offer today :

Offer: iPhone PIN Submit
Traffic Source: Popads
GEO: Thailand
Payout: 1.68$

Results :



Not super promising. 15x payout spent and 2 conversions only, using the top 2 landers that I used for India + 1 ripped. I'll spend another 10x payout on it before I decide what to do.

I also launched my MX campaign today. Unfortunately it's only for 1 carrier and looks like I'm only able to spend 5-10$ a day on it I'll let it run anyway, it probably has low competition so even though the traffic is low there could be a high potential ROI.

Waiting to get my Brasil landers which I should have within 12h. I will launch that campaign as soon as it's ready.

Not much else to report for now. Good luck to everyone

---

Launched the Brasil campaign today, started the India campaign again with the new LP and tried to launch it on Propellerads as well. Launched my MX campaign in propellerads too.

Results for the day :




Ok so first interesting thing that happened when I woke up was that I saw my MX campaign got quite a few impressions. This was surprising as Popads told me I would get less than 1000 impressions per day. Turns out there was some interesting WTF'ness at Popads and it randomly allowed some other cellular carriers besides my targeted one, Fortunately it only spent 10$ before I noticed but a bummer nonetheless. Now when I try to get traffic for my carrier of choice, it tells me there is No available traffic. I then tried to run the same campaign on Propellerads. They have 1000-2000 impressions per day but the CPM to get them is a lot higher (10-15$ CPM). Anyway it's only spending 5-10$ a day so I will see what happens. If anyone knows of a traffic source for pops that has good inventory for Telcel, please let me know

Let's talk about the other campaigns one by one :

Offer : iPhone PIN Submit
GEO : BR
Payout : 0.80 and 0.85$
Traffic Source: Popads

Results :



Considering I was already using some of the best landers from the India campaign, the results aren't very promising. Even by drilling down to the best ROI's the best combo ended up at -35 % :


I decided to pause that campaign.

----

Offer : Amazon Voucher
GEO : UK
Payout : 1.35$
Traffic Source: Popads

Results :

This is a campaign I tried to run a couple weeks ago but I didn't really try that many landers. Now that I found a successful iPhone lander, I modified it to work for a Sweep offer. I made a mistake when setting up the camp and made bids that were way too low and included Desktop traffic. I'm unsure how my CTR's are so low given that in the India campaign my CTR for the same lander is around 45%. Anyway as soon as I fixed the issue I started getting a few conversions. I will re-launch a new campaign with proper settings and let it run over-night and see what kind of results I get. The good news is that if I get OK results, there is potential for scale with this campaign so I'm willing to give it a bit more spend before giving up.

-----

Offer : iPhone PIN submit
GEO : INDIA
Traffic Source: Popads

The best for last, sick 0.28$ Profits

I launched that campaign again last night, split-testing the new optimised lander. Here are the results :


Small sample but the optimised page isn't doing better than the original one, it even has a lower CTR ! Kind of surprised considering I made some speed optimisations and called out user model and brand and changing the wording to be a bit more agressive, making it super clear the user had already won. Not sure what to think about this except that there isn't really any point trying to work on small details on a LP and hoping to increase its ROI that way, at least not until it's already profitable.

I also launched the same campaign on Propellerads, who has way more inventory for my carriers. Unfortunately, after spending 25$ I only got 3 conversions so apparently my campaign doesn't work there. I will gather some more data, specifically placements and carrier data, before I try to launch it on any other traffic source. Since this campaign is pretty much breakeven, I will let it run for a while to gather data on which time's its profitable etc... I've already cut out as much stuff as I can, so I can only spend 10-20$ a day which is a shame since it will take time to get data. Not much I can do about that.
- I just added 3 new landers to this campaign and will be testing them out as well. I'm not super optimistic as I already tried out about 10 landers initially but just gotta keep testing.

Plan for next week:
- Keep testing/optimising the India campaign
- Check the Amazon Offer if it shows any promise
- Talk to my Affiliate managers and ask for more iPhone PIN submit offers.
- Read other follow-alongs, absorb information, read charlesngo blog and read books (Just purchased "The obstacle is the way" by Ryan Holiday)

Thanks for reading

---

Hi, what's your name?

If you haven't already, please take 2-3 recommended AM offers to test. Unless that 1 offer you're testing came out on top. I would then take your best offer and see if other networks have the exact offer or even ask other networks for their top India sweeps offer and test those offers against your winning offer.

It's good that you narrowed down to iOS as you saw Android was doing poorly. Perhaps a strategy you could take is once you got a winning lander and offer then you can open Android targeting back up.

Are you using this tool when it comes to cutting landers?

https://www.peakconversion.com/2012/...al-calculator/

I would leave cutting targeting till the end after you optimized everything like:

- Landing Page Speed
- Best Angle for Lander
- Offers
- Bid Price Points
- Payout Bumps

If after doing all of that you're getting a negative -50% ROI, I myself would ditch the campaign and revisit it at another time. But, that's just me....

I'd hold off on trying different traffic sources until you tried everything to optimize your funnel and perhaps your offer and lander will perform better at a other traffic source. Or your campaign is in green and you can scale to other traffic sources.

I would leave the website quality alone and leave the quality targeting wide open to target all qualities. I would focus more on the targeting factors like OS and Browsers.


If you want you can save a conversation between you and the AM to make sure that in the future if shit hits the fan you have backup. But, I feel like more than likely you can run aggressive with sweeps offers. Usually in my opinion it boils down to the lead quality. If the advertiser likes the lead quality they'll turn the blind eye. Promoting sweeps is far less dangerous than AV.

ROI increase will vary most of the times and I'm sure you're aware of this.

I'd focus on one promotion method now. Master that and move on. Don't jump into multiple things at once.

---

For this you could consider running the campaign during prime hours in India so around 6PM - 12AM in India <----I think that's the prime time. I maybe wrong. Running the campaign during prime hours will allow you to get more people engaged in your ads since they're off work. With more engaged users you're able to get more conversions and then you can find out which lander is your winner.

---

Thanks a lot simon ! The help is much appreciated My name is Klaus

Interestingly, my India offer wasn't performing better at peak times. The hours with the best ROI were 4-6am and 1-3PM or so. 6PM-12AM was about average, with the worse hours being Midnight to 4AM. I only got about 60 leads total on the offer so perhaps it wasn't enough. In any case, I've paused the offer for now as with all the targetting I was getting so little traffic and tried a lot of landers and only managed to get it to -20% or so (which in reality is even worse because the Popads cost were 15% higher than reported in Voluum).

Yupp, this is what I'm doing, I got 4 offers. the Other 2 turned out to do quite poorly and the 4th one was getting very small spend on Popads (about 10$/day) before all the traffic dissapeared on Popads. After 2 days it just said there is no traffic available for my options. I tried to run the campaign on PropellerAds but after cutting all the targetting etc... I was doing about -10% ROI and could only spend 20-25$/day. I decided to pause that campaign and focus on finding other offers that have more potential for scaling.

I am using the suggested tool indeed, very helpful ! I am cutting once I've reached 90% probability.

Thanks ! I will take this approach for my future campaigns.


------

I'm trying to get a hold of my AM's a lot for offers to promote as I don't seem to do well with random offers. The problem is they seem to be quite busy and some of them are in the US which is the opposite time zone for me so we are basically never online at the same time. Not much I can do about that. I re-applied to F5media and Clickdealer in the hopes of getting more offers to promote and split-test. I've had a good pace of launching 1 or more campaigns a day, usually multiple offers.

I'm currently running 3 campaigns. I've tried 5-6 other ones in the last few days but they either got no conversions or did quite poorly (worse than -65% with a proven lander) that I paused them. Here are the 3 campaigns I'm running :

CAMPAIGN 1

GEO : Singapore
Payout : 7.20$ and 8$

7.20 and 8$ Payout. I know yes, it is unwise to try out offers with high payouts at first since it costs more money to test them. However I was out of recommended offers from my AM's and a lot of the iPhone PIN submit offers with low payouts required for me to either apply to run or creative approval. I didn't have much to work on so I decided to go for some offers and this particular landing page only states at the bottom that it costs money. I thought OK why not, I'll spend 50$ on this and only continue if it shows good promise.

Here's the results :



Best Offer : -17% ROI
Offer 1 is outperforming offer 2 with a 93% probability so I cut-out offer 1. Not enough data on the rest to make any conclusions.
My problem here is that I'm getting less than 1000 impressions per day if I remove adult categories (5000 with Adult). However Adult categories performed quite poorly with my India offer. This kind of makes sense. I'm guessing most people watching Porn are not gonna bother signing up up to some iPhone thing, they are kind of busy


CAMPAIGN 2 :
GEO : ZA
Payout : 4$

Again, kind of higher payout but there were 4 different offers across 2 networks that I could test and these fit the profile for my iPhone PIN offers that allow aggro and don't require lander approval.

I just started this campaign so I don't really much data yet. Since I'm using my best lander from the India campaign, I will stop the camp after 5x payout on an offer has been reached without a conversion.


CAMPAIGN 3 :
GEO : Kenya
Payout : 1.60$

Another iPhone PIN offer. I tried 2 bid strategies but still was only able to get a small amount of traffic.
Results :

HiBid Campaign : -37% ROI
LoBid Campaign : +109% ROI



Ok first thing which is weird is that my CTR is over 100%. I've checked my landers for illegal scripts and I can't find any. I've used the landers on my phone and it's not doing anything weird either so I don't really know what's going on here.
Good news is the HiBid campaign isn't doing too bad and the LoBid campaign is positive, although it's spending like 1.5$/day. Just gonna keep running these until I get more data on stuff to cut.
Main problem here is the traffic. I'm now allowed to run Adult Traffic on this campaign so I get like 300 impressions per day. Even if the campaign is profitable I'm gonna make 2$/day or something :/

---
Problems :
- My main issue is getting enough traffic for PIN offers. Since I'm already cutting a lot of carriers and it's only cellular traffic, unless I run a GEO with a ton of inventory, I'm left with little traffic. This is becoming a problem in a lot of these PIN offers. It also means it takes longer to test and there's less things to optimise so I need a decent ROI right away
- I don't have enough offer recommendations. I'm trying to PM my AM's as much as possible but I'm not getting replies as much as I'd like or enough offer recommendations. I'm trying to get into other networks right now to have more potential offers.

Questions :

- I'm currently running all my new offers with my best lander from the India campaign. The lander isn't specific to India, it's just a general lander that I see is often used to every GEO. Now am I correct in assuming this lander will perform well in every GEO, or do some landers work much better/worse in some GEO's than others ? Is it worth it for me to test maybe the top 3-4 landers from my India campaign rather than just the best one ? Similiarly, Adult categories were performing poorly in India, should I expect them to perform poorly in other GEO's too ?

- Since I'm using a proven lander, what kind of ROI should I be looking for to continue a certain Offer ? With a lot of PIN submits I'm only running 3 carriers or so, and it's already all cellular traffic and smartphone only. I'm often running without adult already so there isn't a ton left for me to cut besides OS. If I start cutting OS versions etc... I'm basically left with no traffic.


Thanks for reading everyone !

---

Nice to e-meet you Klaus! -20% ROI is pretty good. Given that India offer payouts are typically low. I looked at your whole post, and one thing I noticed with your strategy is that your blacklisting a lot of targeting factors. I'd suggest before narrowing down the targeting factors is to really work on your offers and landers. Make sure you got the best offer with the best landers. There's nothing you can change on the offers end(or at least that's what I'm thinking, others can chime in on this), however you can always adjust your lander. Let's say you narrowed down to the best lander. You could take your best lander and make adjustments to it. Such as:

- Changing the branding on the page. So let's say you're using a popular brand like Facebook Lander, change it up to a popular brand in India. Just googling some brands in India I saw Reliance Communications as a popular brand. You can create a lander using the same template that you have and create just brand it like Reliance Communications is providing the prizes. Just use discretion when promoting the offer, make sure your affiliate network is cool with this.
- Change up the how your angle. Write a different angle copy.

I'd suggest looking at Amy's post on her case study in Singapore. It could probably help you with this. But as mentioned before try not to cut everything you see not profitable in the beginning instead work on your landers more. You'd be surprised how well adult categories work with sweeps offer. I haven't had luck with popads. But, Adult specific networks work well with some of the sweeps offers I've promoted.

http://stmforum.com/forum/showthread...ght=case+study


- You have to test multiple landers. I used to make this mistake by using a working lander and then porting it to other geos. A lander that could work in one geo may not perform the same in others. So, it's important to rip 5-10 landers to test which one is the best. Then you could use the best one as a benchmark.

- Throw ROI out of the window, it's hard to figure out what type of ROI you should expect. Instead you should focus on what's the real EPV do I see in this geo after multiple rounds of testing. After you figure out what your overall EPV is then in the future you could test what your EPV should be.

---

Hi everyone Been a while but I've been working a lot and testing a lot of different offers, landers and different versions of landers etc...

Got some bad news though today : most likely tore my ACL and will probably need surgery Shit happens.
Good news is that I've got some campaigns doing well and looking promising

Thanks a lot for the help simon !
I followed your advice and decided to not cut targetting until I've tried a ton of landers and lander variations etc...

let's get to it :

CAMPAIGN 1 :
Vertical : PIN Submit
GEO : India
ROI : +17%

Decide to give my India camp another shot. Glad I did.

Here are the results :



Been running it since Saturday evening and there are some good pockets of profit in there. I've narrowed it down to 3 landers which all have a positive ROI.
It's not profitable at all hours of the day so I will start day-parting. Actually during the profitable hours I can get an average of 50%+ ROI. However I just noticed that the profitable hours don't seem to be the same during weekend and weekday. I might run a bit more tests before trying to scale.
Website quality is kind of interesting, not sure what to make of it. Should I start running traffic only above a certain mark or do I need more data ?

Website Quality :



There isn't really anything else I can cut/target (I already dropped Android and tried to put it back with poor results). I did re-add adult sites though and will see if there's anything I can do about that.

Next step : SCALING !
- I will try increase my bid to get more traffic and see the effect. For this, do I set 2 camps and split-tets them again each other ? Unless i choose a really high-bid, the bid with higher camp always "eats up" the volume of the lower camp. Should I run them on seperate days then ?
- I will run the campaign on popunders, tabup, tabunders
- I will try to run it other traffic sources : propellerads, popcash, adcash etc...
- I could add the frequency to 2/24 to see if it's worth it.

---

CAMPAIGN 2 :
Vertical : Sweeps
GEO : Malaysia
Mobile ROI : -20%
Desktop ROI : -28%

These are the stats for the last 3 days and I've since then found some better landers so today the stats are abit better. Mainly Desktop is about breakeven although Mobile took a big hit and is -45% today. I think it performs worse during the week. Need to test more on that.
Both Mobile and Desktop are profitable at certain hours although I am still testing stuff.

PROBLEMS/QUESTIONS :

- For this MY campaign. In Mobile there is a lot of stuff I can cut in targetting. The problem is I don't know which ones to start with/ What order to cut stuff in. Do I cut everything at once, then expand, or cut bit by bit ?. I thought I starting broad, so If half of my traffic is doing -40% vs the other half doing -20%, I would check the -40% camp to see if there is anything green in there, otherwise I would cut it. Not sure how to proceed. Some help would be much appreciated.

Here are the different situations where I could cut something :

Mobile :

Connection :


OS :


Day-parting :


Quality :


So overall, quite a bit of pockets of green. Just unsure what to start cutting first. Maybe start by running only at profitable hours and take it from there ?

The desktop offer isn't quite as targettable but I did a lot better today with some new landers so I will keep it running to see how it does for 1-2 more days and have more data, then start targetting.

-----

OUESTIONS :
- How do you guys test when running bigger traffic ? Say I want to split-test a lander but I only need to spend 10-20$/day to get some initial data on how it compares to a current one. If Im currently spending 100$/day on that campaign in a traffic source, do I just run it for 1-2h or is there a smart way of doing this ? Should I test it on all my traffic sources ? Might Lander A outperform lander B on Popads, but vice-versa on Propellerads ?
- How much do we wanna cut on targetting ? There comes a point where it's not worth it anymore. Is there a rule of thumb for this ?

---

Hey Klaus! Thanks for your PM and for inviting me to your follow-along! And sorry for checking my PMs so extremely late. All kinds of busy lately.

I see that Simon has been doing a simply fantastic job in guiding you! You're in great hands. I see that you've hit green!! Congratulations and a big thank you to Simon for all the valuable insight, and to you for documenting your journey.

Since Simon's already told you everything I would have, I'm not going to interfere too much. Just a few comments about your latest post...


For the IN camp:
-As Simon has pointed out, be sure to use the split-test calculator when cutting landers.
-You can just run the higher website quality levels. Then after you cut down to the winning lander, and/or if you test more offers and find a better one, you can always re-open the other quality levels for a retest. But you're going to do this, then put dayparting on hold, because by cutting quality levels your ROI for all hours will increase.
-You could try scaling now. But why not wait until you have a winning lander and have a higher ROI? Then you can scale using your best targeting, and be sure to test bids at each new source.

For the MY camp:
-Definitely don't cut everything at once - for the same reason mentioned above: Often when you cut one thing, ROI for other traffic segments will increase.
-Also - are you tracking the [ISPID] and/or the [ISPNAME] tokens from popads? Would be good variables to drill down to.
-Mobile carrier would be another good one to drill down to.
-There are many ways to do the cutting. I would place the most importance on the major OSs - android and ios - since they are basically the most major traffic segments. I would use the current camp for android, and start a separate camp for ios. I would cut for EACH of ios and android. For each OS, I would start by drilling down to mobile carrier and [ISPID]/[ISPNAME] to see if anything performs terribly and start by cutting those. Cutting by ISP on popads is tricky - will need to see stats to provide more advice.


Both camps look promising! Great work guys!



Amy

---

Thanks Amy !

I've spent the last few days just gathering data and tried out a few new landers. Unfortunately, I haven't found any landers better than my current ones and I've probably tried at least 20 of them total, so I don't think I will find anything much better with ripped landers. Should I hire someone to make me custom landers ? These are quite expensive so I'm not sure what I should ask for. A variation of my current ones or something completly new ?

I've also tried a couple different angles but I haven't noticed any significant increase/decrease in ROI with different angles.

Thanks. I managed to hit green on both camps and tried to scale to other sources but unfortunately couldn't really hit green anywhere besides some promise at ZeroPark and Adcash. A lot of the other traffic sources don't offer targetting as good as Popads. I'm gonna focus on cutting down targetting and trying out more bids for my campaigns while trying to brainstorm new angles and what kind of landers I would like to be made for me.

I've racked up over 1000 conversions now for the Malaysia offer for the past 10 days or so, should I ask for a paybump ? India offer also been running 2 weeks with about 200 conversions, is it reasonable to ask for a paybump ?

---

When optimizing any one variable, you'll inevitably get into decreasing rates of return sooner or later. Sounds like it would be better to test offers instead of landers!

Yes you can try completely original landers, and you're encouraged to do so. But personally I like to leave that until (IF) I find a promising offer - after I've already scaled the camp so that any increase in ROI I can achieve with a new lander will result in significant increase in profits.



Yup - not every traffic source will work. But the better your funnel is, and the better you know what target works, the better your chances will be when scaling to other networks.

But most important of all, I'd recommend that you test more offers!



Absolutely! Ask for a quality check first, then when they tell you your quality is good - that would be the time to pop the request about a pay bump!




Amy

---

Home > Mobile > Follow-along Campaigns