Home > Push Traffic >

Dating and Push to consistent profits (28)


03-29-2019 11:22 AM #1 mstgeorge (Member)
Dating and Push to consistent profits

Hi guys,


The last couple of months I am running casual dating with push traffic and I have mixed results, some green camps some failed and some breakeven.

I have been struggling with lead quality and traffic sources as you can see in the thread I have posted here:

https://stmforum.com/forum/showthrea...-casual-dating

I also have problems with consistency and still haven’t built a system to steady profits.


I am starting this FA to share my process of building a campaign and ask questions at every step in order to get help and see where I am doing mistakes.

My target with this FA is not to reach xx$ or xxx$ per day but to build a system that will allow me to reach consistently successful campaigns.

A little bit about the setup:

-Ad Networks: Clickdealer, Xcash, UnitedGap, Imaxcash (Do you have any other networks with casual dating offers I should apply?)

-Traffic Sources: Admaven, Zeropark, Propeller (Again any suggestions are welcome)

-Testing budget: 30$ - 50$ per day

So where are we right now:

Offers

Testing 4 offers from 3 different networks

GEO:ES
Platform: MOB
Payout: 1,4$ -1.65$

Landers

We have 8 landers ripped and translated with different styles 7 step quizzes, 3 step quizzes, comic quizzes, 1 step etc.

For the current campaign, we started testing 3 of them and will continue to test more when we find winning offer.

Creatives

Started testing 3 different angles that copied from spytools and waiting to find a winner and try different variations of it.

So the first step I take when I launch a new campaign is to try to find an offer that seems promising and test landers and creatives with it. In order to do so, I use the 10x payout rule to eliminate offers and if an offer reaches 2 conversions before the 10x payout I pause the others use it for testing.

Do you think it is a good approach to start?

I just launched the ES campaign a few hours ago so I will update with some stats in a couple of days.

Thanks


04-01-2019 09:51 AM #2 mstgeorge (Member)

Update


So I started to run the campaign 2 days ago and here are the stats I have by now:


Cost: $35,34
Revenue:$7.45
Profit: - $27.89
ROI: -78.92


What I have done till now:


-In the first day one offer got 2 conversions out of 3 total so I paused the other 4 offers and continued testing landers with this


-The second day I put the stats of the landers performance at the bayesian calculator tool (
https://www.peakconversion.com/2012/...al-calculator/) and found out that one of the 3 landers had only 5% chance of being the best so I cut it and let the other 2 running.


-Also killed 2 placements that had no conversions and 2x plus the offer spent


Questions


-Do you think I did good to pause the rest of the offers or should I have left it a bit more?


-This camp doesn’t look really promising should I leave it for the day and kill it tomorrow if something doesn’t change dramatically or give it a bit time?


04-01-2019 10:23 AM #3 maynzie (Moderator)

Whats up buddy!

-Do you think I did good to pause the rest of the offers or should I have left it a bit more?
I feel it was probably a bit early to cut offers based on that spend and that many landers there is probably a bit too many variables going on, maybe have left for another day but I am not to up to date with the best casual dating offers so hopefully someone else can chime in here and shed some light on the best networks. Is there any offers you can find in the $2-3 mark?

I would have in the beginning tested 3-4 offers like you've done, 2-3 landers (all different types), and 5-6 creatives.

-This camp doesn’t look really promising should I leave it for the day and kill it tomorrow if something doesn’t change dramatically or give it a bit time?

Cost: $35,34
Revenue:$7.45
Profit: - $27.89
ROI: -78.92
Thing is yes from that perspective it doesn't look too promising but there are so many variables bro, what happens if you spent another $100 and an offer began to stand out with a certain type of lander you could then begin to tweak on, not to mention creatives and which style to pursue. And lastly its push traffic, optimising the site ID's makes a big difference too, its far too early to make any judgement on the campaign in the regard of whether or not it will work but I think you're just probably testing too much at once.

Lets see how day 2 goes with the adjustments you did and see if it gets any better or anything stands out.


04-01-2019 01:34 PM #4 matuloo (Legendary Moderator)

I have been struggling with lead quality and traffic sources as you can see in the thread I have posted here:
Push traffic is like that, the quality is not backing up for all advertisers. Lot's of dating is running on push, but you simply have to expect some quality issues. It can work though

-Ad Networks: Clickdealer, Xcash, UnitedGap, Imaxcash (Do you have any other networks with casual dating offers I should apply?)
I would also add advidi, togethernetworks ... pretty much all networks have some dating, and you never know where a gem is hiding

-Traffic Sources: Admaven, Zeropark, Propeller (Again any suggestions are welcome)
richpush and megapush can also work with dating

-In the first day one offer got 2 conversions out of 3 total so I paused the other 4 offers and continued testing landers with this
This was a bit too soon indeed, you need more data to confirm an offer is really a winner, you want it to stand out more so to speak.

-This camp doesn’t look really promising should I leave it for the day and kill it tomorrow if something doesn’t change dramatically or give it a bit time?
ROI looks bad for sure but as maynzie mentioned, there are quite a few variables you can mess with. You can come up with more ads too, those can make or break a campaign too. Do not underestimate the power of an AD. I think I had the best results with dating on push traffic from propeller and now they allow multiple ads in a campaign so take advantage of it.

And you can always test another GEO too, spain isn't really my expertise so not sure how well it can perform, there are certainly better GEOs out there for dating. Other EU contries, Nordics ...


04-01-2019 02:15 PM #5 mstgeorge (Member)

Hi Maynzie thanks for jumping in,

I feel it was probably a bit early to cut offers based on that spend and that many landers there is probably a bit too many variables going on, maybe have left for another day but I am not to up to date with the best casual dating offers so hopefully someone else can chime in here and shed some light on the best networks. Is there any offers you can find in the $2-3 mark?
Unfortunately i didn't found any offers with a better payout from the ones i mentioned above.

Thing is yes from that perspective it doesn't look too promising but there are so many variables bro, what happens if you spent another $100 and an offer began to stand out with a certain type of lander you could then begin to tweak on, not to mention creatives and which style to pursue. And lastly its push traffic, optimising the site ID's makes a big difference too, its far too early to make any judgement on the campaign in the regard of whether or not it will work but I think you're just probably testing too much at once.

Lets see how day 2 goes with the adjustments you did and see if it gets any better or anything stands out.
Makes total sense i will gather more data and see how it plays out.

Thanks


04-01-2019 02:42 PM #6 mstgeorge (Member)

Hi Matuloo,

Thanks for the suggestions!

-In the first day one offer got 2 conversions out of 3 total so I paused the other 4 offers and continued testing landers with this
This was a bit too soon indeed, you need more data to confirm an offer is really a winner, you want it to stand out more so to speak.
Probably i messed up when i wrote the post and didn't explained properly my strategy when testing.
When an offer seems to stand out (by making the 2 conversions) i pause the others and use the "winning" offer to find the best lander, after i find the winning lander i start running the paused offers to find a winner.
Do you think is a solid strategy for push or i should change?

ROI looks bad for sure but as maynzie mentioned, there are quite a few variables you can mess with. You can come up with more ads too, those can make or break a campaign too. Do not underestimate the power of an AD. I think I had the best results with dating on push traffic from propeller and now they allow multiple ads in a campaign so take advantage of it.
Also about giving up on a camp. What i use as a general "rule" is when an offer has spent 10x payout * different variables (lander, creatives) and doesn't look good ROIwise to kill it. What do you think of this strategy? Do you have a different approach on this as an experienced marketer?
Any help is much appreciated

Thanks


04-01-2019 08:05 PM #7 datingfactory ()

We own some casual and adult dating offers for ES at GeoEarnings.com. We would be happy to get you setup and test things out.


04-01-2019 08:41 PM #8 matuloo (Legendary Moderator)

Probably i messed up when i wrote the post and didn't explained properly my strategy when testing.
When an offer seems to stand out (by making the 2 conversions) i pause the others and use the "winning" offer to find the best lander, after i find the winning lander i start running the paused offers to find a winner.
Do you think is a solid strategy for push or i should change?
I know what you mean, but 2 conversions simply isn't enough, let me explain why. Adult is an impulse vertical, so sometimes a few conversions can be random ... you can simply hit a guy who'd signup to literally anything, simply because he's in such mood (after looking at all the porn . It's possible that one of the 2 leads was such a guy, so you actually based your decision on 1 lead. Know what I mean? So in order to be certain, you need an offer to stand out more than for example 2-1-1-0 (the numbers should represent 4 different offers). I f you want to rush it , wait for at least 3, but preferably 5. Or look for very clear patterns, like 3-0-0-0 or maybe even 2-0-0-0 but don't try to pick a winner based on 2 leads if there are also other offers that converted. I understand you just wanted to quickly find a solid offer so you could move to other parts of the funnel, but still, keep the randomness in mind.

Also about giving up on a camp. What i use as a general "rule" is when an offer has spent 10x payout * different variables (lander, creatives) and doesn't look good ROIwise to kill it. What do you think of this strategy? Do you have a different approach on this as an experienced marketer?
The key is to give each combination at least 1x spend, 2x is better. And to save some $, you can cut the worst combinations earlier when you see they perform way worse than the rest. So for example, if you see that some combinations already made a conversion, but some did not and on top of that, they have low LP CTRs or the respective ads have very bad CTRs... chances are they simply won't perform and you can cut them.

The ideal scenario I'm looking for when starting a new campaign is to see some combinations either stand out, so I can keep just those and give them more traffic OR when some start to clearly fall behind, so again I can keep a limited amount of them and give them more data. If everything converts on a similar level, but well below the breakeven point thats a bad sign and I pause such campaign as it clearly wasn't a hit.

We also need to set one thing straight : when a campaign doesnt perform well, you don't have to stop it for good, throw out the offers and move to other GEO or something. In many cases it's the funnel that's causing the problem and the offer can work fine. Or it might be the traffic source or some part of the targeting. I give every offer several tries, before completely ditching it. Unless of course I have a well working funnel in the same GEO/targeting at that point. If a new offer isn't working with a strong funnel, then obviously I dont mess with it for too long.


04-02-2019 02:32 PM #9 mstgeorge (Member)

Thank you very much Matuloo for the detailed answers, I think that now I understand better about the importance of gathering some data before cutting anything especially in the beginning and readjust my strategy. Also it became more clear what to take into consideration when deciding if you should kill a camp.


04-02-2019 03:01 PM #10 mstgeorge (Member)

Update



As I was advised by Maynzie and Matuloo I think I was trying to rush things a bit so I let the camp run to gather some data.


Cost: $56,16
Revenue:$18.65
Profit: - $37.51
ROI: -66.79%


Better than yesterday but again not looking so promising...


Here are all the changes I have madein the meantime


Landers


Almost all the conversion came from one of the two remaining landers so I plugged the data I have gathered to the bayesian tool and saw that the bad performing lander had only 5% chance of being the best so I cut it. So we have a winner in this part of the funnel :).


Offers


Since I had a winner on the lander front I unpaused the other 4 offers and startedsending running traffic to them. Now i wait to gather some data to see if I can find an offer that stands out.


Creatives


From the 3 banners I have been runningalmost all the conversions came from one, again I plugged the data to the kill-whitelist calculator and found that I have to kill other 2. Then took the winning banner and created 2 variations of it. At the first one I used the same angle with a different image and at the second I kept the image of the girl and changed the angle.


Placements


Cut 4 placements that have spent over 2x the offer payout with 0 conversions.


What I learned till now


As I mentioned above from the answers of Maynzie and Matuloo I understood that I was rushing a bit on cutting stuff so I am readjusting my strategy and will try to gather some data before I start cutting offers.


About when to give up a campaign it became very clear what I have to do in order to decide to kill it or not.


After reading the responses of the most experienced members here and digging a bit into the forum I understood that I didn’t gave to the creatives the attention they need. What I did till now is I to find some creatives from spytools and throw them into the mix and wait to see some result.
I am readjusting my strategy in this part too and I will make more due dillingence about angles, copy, images etc. before I start a campaign and also test more creatives in the process of testing.


Questions

- The mindset I had till now was to test as many offers as I can in order to find a profitable one. So I didn’t test other factors too much, just tested a sh***load of offers. As I understand from the above replies at push (or in AM in general?) isn’t much like that. It’s not about testing a ton of offers but more about building a solid funnel during testing phase and use that funnel to see if the suggested offer works for you. I don’t know if that’s make sense but what do you guys think?


04-02-2019 04:09 PM #11 maynzie (Moderator)

- The mindset I had till now was to test as many offers as I can in order to find a profitable one. So I didn’t test other factors too much, just tested a sh***load of offers. As I understand from the above replies at push (or in AM in general?) isn’t much like that. It’s not about testing a ton of offers but more about building a solid funnel during testing phase and use that funnel to see if the suggested offer works for you. I don’t know if that’s make sense but what do you guys think?
The key is to give each combination at least 1x spend, 2x is better. And to save some $, you can cut the worst combinations earlier when you see they perform way worse than the rest.
It's a mix of everything though, over time you become more intuitive (especially to master matuloos comment above) to which landers work but thats from testing a lot at the start, but yes a good offer makes or breaks campaigns in most cases. You can have a really good funnel that is doing so well with an offer, then suddenly its paused and you try to find another one but nothing will match that same ROI. Funnel is something that "you can control on your own", offers is more you have to network and research to discover or build relationships over time - a little more of a curveball really.

Nice to see your ROI rised slightly, and to see more conversions. Sounds good on cutting down to the one lander, now keep testing the offers next phase run the best one and then continue to wash in new creatives to test against your best ones. Also remember its push traffic, you have ability to cut the fat of non converting targets later too - thats a big part still in play in your optimising. ROI's like you have now are no way dead campaigns, and for educating purposes I still feel like you'll see much more improvement in the ROI however I'm no expert in Spanish dating on push so its hard for me to comment on it. I feel like the payouts may be a little too low....


04-03-2019 10:04 AM #12 matuloo (Legendary Moderator)

- The mindset I had till now was to test as many offers as I can in order to find a profitable one. So I didn’t test other factors too much, just tested a sh***load of offers. As I understand from the above replies at push (or in AM in general?) isn’t much like that. It’s not about testing a ton of offers but more about building a solid funnel during testing phase and use that funnel to see if the suggested offer works for you. I don’t know if that’s make sense but what do you guys think?
As Maynzie mentioned already, it's a mix. You need a solid offer, no questions about this, it's the most important part of the puzzle. But without the right funnel, it's still not gonna work properly. So you need to find the right LP and ADs too. At the beginning it's hard, because you are not sure about any part of the mix. You don't know if the offer is solid, you don't know if the LP is solid and you also don't know whether you're using the right ADs. So the only way to do it, is to start with something and then optimize the other parts one by one.

You have an LP that converted multiple times, so even though we can't say yet that it's a really good one and possible to make it profitable, you can still use it as a benchmark and base your future tests on it. It is converting, so that will do for now. Now it's time to move to the next step, you mentioned that you added some new offers, so let offers be the next tested element. Wait for the data and post the results here. In an ideal case, there will be one or two that outperform the rest so you will choose those and then you can move to the next part, which are the ADs. Give them some data, pick the best ones and you will end up with a full funnel that will serve the purpose of a starting benchmark.

When this initial testing phase is over, you will have a set of ads, LPs and offers that are working to some extent. And from that point on, you will focus on the particular parts of the campaign, and try to make it better. You will try to find an offer that beats the benchmark, then an LP that will beat the current top performer, then the AD ... and in the meantime, you can keep on cutting placements that simply underperform. Under a few rounds, you should find something profitable.


04-05-2019 04:43 PM #13 mstgeorge (Member)

Hi guys,


Thanks @ Maynzie and Matuloo for the valuable responses. They are really useful and help me make a plan on how to act when testing a campaign.


There are a lot of updates so I will cut straight to the chase.




Cost: $134,82
Revenue: $37.2
Profit:- $97.62
ROI: -72.41%


So, since the last update I am running traffic to the offers and trying to figure out the best one, I used a tool (https://win-vector.shinyapps.io/CampaignPlanner_v3/) to evaluate which offer to stop but seems to be down the last couple days so here are the stats of the offers:





What I have done so far is to cut the offer in the box. It seems that the one with 7 conversions, stands out but are these enough data?


I think Maynzie was 100% right when he said I am testing a lot of variables and I understand it now seeing how long it takes to eliminate the bad performing ones.


Placements


I have been cutting lots of placements that underperform during the process using the kill whitelist calculator, but there is a specific one that I am not sure what to do. I am talking about the first row in the screenshot below.


Kill whitelist calculator tells me to kill it, also the CPA is more than double the offer payout. On the other hand more than half of the traffic and conversions come from it. Do you think I should follow the tool and ditch the second thoughts I have?




Creatives


Again the one that was performing better still keeps the lead. I changed the other 2 with some completely different angles I found from my new spytool.
I haven’t posted stats of my creatives till now. The CTRs range from 0,10 – 0,40 I know that there are not standard numbers but do you think I am in the “normal” zone or I have to completely change?


In the meantime I have started 2 more camps so why not ask some stuff about them, right?


Lets call them campaign 2 and campaign 3.


Campaign 2


This is a pretty fresh campaign running a little over 24 hours the Geo is GR and I test 3 DOI offers with 3 landers and 4 creatives and running at propeller. Forgot to mention the payout is $2.3


Cost: $33,5
Revenue: $0
Profit:- $33.5
ROI: -100%



Not much to say here weird thing about this campaign is that the stats from creatives and landers are really good. The creatives CTR are from 0,4 to 1,2 and the landing pages CTR from 42% to 64%!


I have sent about 500 clicks to the offers and no one converted which looks pretty strange to me, I even tested myself the offers did a conversion and asked the AM if everything is ok.


I am thinking to kill this campaign but the only reason I am letting it run is that it is a DOI and I haven’t run a DOI before so I am guessing it might need a bit more data. Do you think I am burning money with this one?




Campaign 3


This camp is also running almost 24 hours Geo is Italy, 2 DOI offers, 1 SOI again 3 landers 4 creatives traffic source is propeller and payout is $1,5 to $1,7.


Cost: $46
Revenue: $14,6
Profit:- $31,4
ROI: -68,26%



Offers


The 9 conversions are spread 5-2-2 with about 100 views each one, the 5 is the SOI offer. Do you think I should cut the other 2 or collect more data? Keep in mind that SOI has a better payout than the DOIs.


Landers


Things were a bit more clear here. From the 6 conversions at the moment I checked it, 5 came from one LP.


I paused the other 2 and let it run.


Creatives


There was no clear winner here all the conversions were spread evenly so I didn’t do anything.




Questions


Campaign 1 Spain


-Should I stop some more offers or let them run and collect more data?
-Do you know any other tool like the one I mentioned that I can use to evaluate offers?
-What do you think about the placement I mentioned above should I keep it or blacklist?
-What do you think about the creatives stats are they “normal” or I should change them?


Campaign 2 Greece


- I think there is only one question here, should I keep it or not?


Campaign 3


-Should I cut the 2 DOI offers and keep the SOI?


General questions


-About DOI offers, do I need to give them more time? Conversions arrive late cause people might open their mails later?


-About day-parting, I think I see some dead hours. Probably people sleeping and notification get buried under other stuff or something. Is it something that you consider optimizing in the first part of testing phase or do it after you have a funnel?


-Do you think I am spreading myself thin among 3 campaigns? I am doing affiliate full time now so I think I have the time to manage them. What do you think?


- I am optimizing a lot of things at the same time, what I mean is that when I see something underperforming I cut it. You think I should prioritize my process ex. Offers--- landers---- creatives as Matuloo mentioned above even if I see that something isn’t performing well?


04-07-2019 08:21 PM #14 matuloo (Legendary Moderator)

What I have done so far is to cut the offer in the box. It seems that the one with 7 conversions, stands out but are these enough data?
It's not THAT much data, but the numbers speak quite clearly, the offer with 7 conversions really stands out. This is actually the best case scenario, when one offer stands out like this, focus on it and try to optimize it into profit.

Kill whitelist calculator tells me to kill it, also the CPA is more than double the offer payout. On the other hand more than half of the traffic and conversions come from it. Do you think I should follow the tool and ditch the second thoughts I have?
This is the question, right? Sacrificing the largest spots literally "hurts", but if they don't perform, what sense does it make to keep losing money on them? What I'm doing in cases like this, is to pause them, then work on the funnel and once its stronger, I try to relaunch these large spots.

The CTRs range from 0,10 – 0,40 I know that there are not standard numbers but do you think I am in the “normal” zone or I have to completely change?
CTRs can really vary a lot, when I look at my dating campaigns on push, the CTRs go all the way from 0.1 up to 5% and even more in rare cases. Really hard to set a "standard" number here.

I am thinking to kill this campaign but the only reason I am letting it run is that it is a DOI and I haven’t run a DOI before so I am guessing it might need a bit more data. Do you think I am burning money with this one?
$33 spend on a $2.30 payout is more than enough. Since you didn't get any lead, it needs to get paused for sure. There is either some problem with the offers or you're getting bot traffic, or the funnel totally sucks ... or the funnel is exceptionally bad. But either way, this campaign won't work as is.

The 9 conversions are spread 5-2-2 with about 100 views each one, the 5 is the SOI offer. Do you think I should cut the other 2 or collect more data? Keep in mind that SOI has a better payout than the DOIs.
5-2-2 is pretty strong argument in favor of the SOI. Especially when the SOI has a higher payout. So if you wanna continue testing, go with the SOI and maybe look for better offers too.

Campaign 1 Spain


-Should I stop some more offers or let them run and collect more data?
-Do you know any other tool like the one I mentioned that I can use to evaluate offers?
-What do you think about the placement I mentioned above should I keep it or blacklist?
-What do you think about the creatives stats are they “normal” or I should change them?
I answered most of the question, there is just the "ads" to address: you can pretty much always come up with a stronger ad, so definitely keep on testing more.

-About DOI offers, do I need to give them more time? Conversions arrive late cause people might open their mails later?
Late conversions are common with DOI, but it's not like you will get a ton of them. Let's say you send 10 DOI leads, so chances are on the next day, there will be one more. Sometimes 2, sometimes none. Late leads will not turn a heavily losing campaign green, consider them as a bonus.

-About day-parting, I think I see some dead hours. Probably people sleeping and notification get buried under other stuff or something. Is it something that you consider optimizing in the first part of testing phase or do it after you have a funnel?
You need a lot of data in order to spot dead hours and to be able to daypart, this is one of the last things to try.

-Do you think I am spreading myself thin among 3 campaigns? I am doing affiliate full time now so I think I have the time to manage them. What do you think?
As long as you feel comfortable about this and you have the budget available, 3 campaigns is not a lot at all.


04-08-2019 04:34 PM #15 mstgeorge (Member)

Thank you matuloo for the answers.

So let’s see how I am doing


Campaign 1 Spain


Cost: $222,60
Revenue: $96
Profit:- $126,6
ROI: -56,87%



Things finally seem to be improving, it’s been 3 days since I cut the offers and the placement I asked above and my ROI has improved dramatically it’s -30% for these days and even one day was in the green


So right now I think I am really close getting to breakeven, the things I have done these last days besides cutting offers and the placement I asked about are:


-Split test a completely different landing page (sent 30% of traffic there) but it clearly fell behind the winning one
-Created a variation of the winning landing page and put it in rotation after the above failed(haven’t gathered data yet it’s couple hours ago I put it)
-Paused 2 out of 3 creatives that had to be killed according to kill whitelist calculator and created other two that I am testing till now
-Continued cutting placements that underperformed, drilled to my tracker trying to find some segments that can be cut (OS, Devices, Browsers etc.) with no luck as all segments were having similar stats.

So what do you think my next steps are? Focus on the ads and try to test different angles and creatives ? I think stoping the camp is not an option right now right?


Sorry for asking this I know you guys don’t like giving exact numbers or rules. When you pass the test phase 1 and have a benchmark that somehow works, like the stage I am now, do you have a rough number for ROI to decide to stop or not the campaign?(especially if something doesn’t stand out in the segment area.)


Campaign 2 Greece


Paused it for now.


Campaign 3 Italy


I have some questions about this campaigns that it hasn’t to do with ROI etc.
I started running this campaign at propeller with a testing budget of 20$ per day.
After about 48 hours of running I noticed at the stats that all the daily budget was spent at the first two hours of each day. I talked with support and I was told that my options were to increase the frequency cap and/or decrease the bid.


So do you think I have accurate statistics for this campaign considering all the data have been gathered in two hours?


Do you think decreasing the bid is a valid option? I mean I won’t be getting traffic from all crappy placements if I do so?


How do you deal when having a small testing budget with a situation like this?


04-08-2019 09:14 PM #16 matuloo (Legendary Moderator)

Things finally seem to be improving, it’s been 3 days since I cut the offers and the placement I asked above and my ROI has improved dramatically it’s -30% for these days and even one day was in the green
Great to hear that! Seeing a green campaign at last must have felt great, am I right? Now you know it can work!

So what do you think my next steps are? Focus on the ads and try to test different angles and creatives ? I think stoping the camp is not an option right now right?
Yup, testing ADs and creatives should be your focus now. You have a somewhat solid funnel now, so you have something to compare the new variations against. Don't forget about the offers too, there are certainly different offers out on the market, try to look for them. And even if you don't find some straight away, most offers come with multiple urls so try to test more of them too. Quite often, they can perform very differently.

Sorry for asking this I know you guys don’t like giving exact numbers or rules. When you pass the test phase 1 and have a benchmark that somehow works, like the stage I am now, do you have a rough number for ROI to decide to stop or not the campaign?(especially if something doesn’t stand out in the segment area.)
Yup, it's not like we wouldn't want to, it just depends on so many factors and every campaign is different. So giving exact numbers can be very misleading. But ok, you asked for it Once I know something about the market and have some tested creatives for it ... I don't like to see when a new campaign in such markets goes below the breakeven point much. A somewhat proven funnel shouldn't go negative to much as there isn't THAT much space for optimization left. -20% is optimizable, but when I see something like -50% I know something is wrong. On the other end of the spectrum, as long as I can make +10% ROI I keep the campaign running, unless it interferes with my cash or work flow. Of course I aim for more, but I won't say to any possible profit.

So do you think I have accurate statistics for this campaign considering all the data have been gathered in two hours?
Italy is a large GEO so yes, it can happen.

Do you think decreasing the bid is a valid option? I mean I won’t be getting traffic from all crappy placements if I do so?
Are you targeting the higher quality traffic with propeller? If not, try that. You're right that the clicks from crappy placements are often sold at a lower bid, but unless you have them blacklisted, you will get traffic from them even with a high bid Lower bid can mean lower quality traffic, but it's not a must. Even here on the forum, many newbies reported better results with lower bids and quite often it was with propeller and their POP traffic. I wouldn't be afraid to try it.

How do you deal when having a small testing budget with a situation like this?
In a high traffic GEO you just have to limit the targeting somehow ... drop an OS, use tight capping, lower the bid, target just one quality level ... basically just limit the target group by any means possible, in order to reach balanced distribution, even if from lower amount of targets.


04-11-2019 06:39 AM #17 mstgeorge (Member)

Thank you matuloo once again for the answers.


Things didn’t go as expected for my campaigns…


Campaign 1 Spain


Cost: $265,64
Revenue: $105,6
Profit:- $159,04
ROI: -60.25%



After the last update the campaign performance dropped significantlyIt is around -70% and I cannot improve it no matter what I split test 2 landing pages sending a small potion of the traffic I changed creatives (4 new) played with the bid but no improvement so I decided to pause the campaign.


In total I have tested 6 offers 5 landers and 9 creatives


Do you think I shouldn’t have stopped it and try more? and if so, what should I have done?


Campaign 2 Italy


Cost: $146.16
Revenue: $43.2
Profit:
- $102,96
ROI:
-70,44%



I have started with 3 offers and 3 landers and 4 creatives and using all the tips mentioned above I have cut down to one offer and one lander. I have changed the creatives and changes 4 more with no improvement. So I paused this campaign too. After the optimization and creatives testing it is stuck at around -50% and I don’t think I can make it green.



After pausing these 2 campaigns I have started a new one in France with one offer using a lander that performed the best at all my recent campaigns and 4 creatives. It has been running for a couple of hours with some conversions but I wait to gather some data to post here.



Also setting up another campaign at Poland with 3 offers starting with the best performing lander mentioned above and 3-4 creatives.





Questions



-Did I do right to pause the campaigns or not? (If you want some stats from the tracker please tell me to post)
-Also is it a valid strategy when starting a new campaign to use a proven lander from other campaigns and test more landers later?


Thanks


04-11-2019 10:58 AM #18 matuloo (Legendary Moderator)

In total I have tested 6 offers 5 landers and 9 creatives


Do you think I shouldn’t have stopped it and try more? and if so, what should I have done?
Just to give you an idea, I tested several 100s of banners, dozens of LPs and at least 20 offers, before I landed my first stable profitable campaign. So you just scratched the surface pretty much

But I also switched GEO as I couldn't make it in the first one, so hard to say whether you should have sticked with Spain or not. You didn't run a lot of tests, that's for sure.

-Did I do right to pause the campaigns or not? (If you want some stats from the tracker please tell me to post)
Your ROI didn't look good, that's a given. But you were receiving conversion so I'm sure you could actually improve with better creatives ... Sometimes it's just about coming up with the right angle for the creative. I'd give it more time.

-Also is it a valid strategy when starting a new campaign to use a proven lander from other campaigns and test more landers later?
Yes, because that way, you at least know that the lander is somewhat solid. So you have some kind of a benchmark that you will try to beat. Just keep in mind that even the best LP you have now, is probably not the real winner, so you still need to improve it ... with better copy, better angle, better imagery etc ... But generally speaking, I also always start a new campaign with lander that performed well in other campaigns.


04-12-2019 03:54 PM #19 mstgeorge (Member)

So as I mentioned above I paused the ES and IT camps for now I might test them again after I improve my funnel but for now I will move on.


So I have started 2 campaigns and I am in the process of creating a third


Campaign 1 France


Cost: $54,3
Revenue: $62,4
Profit:
$8,10
ROI:
14,92%



So this campaign started really good I am running it a bit over 48 hours.


It has one offer and started with a somewhat proven lander from previous camps & 4 creatives in rotation.


Haven’t done much, just made a variation of the lander (decreased the steps) and send it 30% of the traffic.
About the creatives i see one that is converting better than the others till now as you can see below.
What's my strategy here?
Create variations of it and test 'em against it?
Also should i go crazy with testing or just test couple banners each time against the good converting one?





Campaign 2 Poland


Cost: $22,38
Revenue: $9,32
Profit: -$13,06
ROI: -58,36%



Here I started with 3 offers again the best performing lander and 4 creatives not too many stats to do something I just cut one offer that had 0 conversions while the other two had 4 each. Just waiting to gather data and see.


Questions


When you have green camp and want to test different creatives, landers etc. You do it by testing one element at a time or a lot?
What I mean for example If I want to test several landers against my winning one I put a bunch of them in rotation or test one by one sending a small portion of the traffic?


04-15-2019 11:20 AM #20 mstgeorge (Member)

A small update of what happened in the weekend

Campaign 1 France



Cost:$186,6
Revenue:$209,3
Profit:$22,7
ROI:12,17%

It feels good to run a campaign that is not losing you money.
So what I did in the weekend was:

-Stop the original lander as the variation (less steps) was performing better.
-Tested some variations of the winning creative and atm have 2 similar creatives running. (same copy different icons)
-Cut some big placements that were performing bad.
- Increased the daily budget at $50 and then $75.

Yesterday the campaign took a huge dip and went from +60% (previous day) to-30%. I have read at several posts here at STM that it is pretty normal for camps to have these kind fluctuations especially if you run small budgets like me, but of course I have some second thoughts that might banner blindness have kicked in so I decided to ask here

Campaign 2 Poland

Cost:$75,38
Revenue: $31,07
Profit: -$44,2
ROI:-58,77%

-Cut the other offer and running 1.
-Added 1 variation of the proven lander but killed it as it didn’t perform well.
-Killed 2-3 placements
-Have tested 11 different creatives 4 of them running right now.

Campaign 3 Malaysia

Cost:$48,5
Revenue: $15,3
Profit: -$33,2
ROI:-68,48%

This campaign is running for 3 days.
It started with 2 offers, 1 lander and 4 creatives.
-I have cut one offer and again added a variation of the proven lander. Still collecting data from the landers.
-I started with 4 creatives cut 2 and added other 2 to test.

At this point with campaign 1 I think I need to improve my ROI and then start thinking about scaling etc. I am testing some creatives and also planning to test couple more landers and also add another offerI found.

Do you think my plan is solid?

For the second and third campaign I am stuck at -50% -60% range.
I think my main mistake in the past was not testing so much creatives so thats what I am gonna do right now.
In the past 2-3 days I have tested several creatives but saw no improvement so I will test more in the next days and see what I will do.


04-15-2019 12:52 PM #21 matuloo (Legendary Moderator)

Create variations of it and test 'em against it?
Also should i go crazy with testing or just test couple banners each time against the good converting one?
Yup, especially in case you are certain it's the ad that makes the difference. But there are two more that are quite close. Do they have something in common maybe? There might be some trends that tells you what to focus on.

No need to get wild too much, but don't slow yourself down to much either. ADs in push die quickly so you gotta move fast.

When you have green camp and want to test different creatives, landers etc. You do it by testing one element at a time or a lot?
What I mean for example If I want to test several landers against my winning one I put a bunch of them in rotation or test one by one sending a small portion of the traffic?
You can test multiple ones, just don't test all parts of the funnel at once. If you want to test ads, feel free to test multiple, but don't change LPs at the same time, for example.

Yesterday the campaign took a huge dip and went from +60% (previous day) to-30%. I have read at several posts here at STM that it is pretty normal for camps to have these kind fluctuations especially if you run small budgets like me, but of course I have some second thoughts that might banner blindness have kicked in so I decided to ask here
Fluctuations are normal yes, but it won't hurt to throw in a different AD to test it or even start a new campaign from scratch, even if you use the same funnel. The push algos act weird at times, to sometimes a newly created camp with the same funnel can work better.

Are you also cutting the poor placements? That can help a lot too.


04-16-2019 08:00 AM #22 mstgeorge (Member)

And once again i was paused from the FR campaign for quality issues.
It is really frustrating having this happen once more so i took the decision to stop running dating with push.
I know a lot of people are running dating with push and making tons of money but for me doesn't work as i always have problems with lead quality.
So i have a decision to make here and i wanted to ask the opinion of more experienced people here.
Should i switch traffic source?(run adult dating with banners)
Or switch vertical?(run sweeps with push)


04-16-2019 08:40 AM #23 AdMaven (Veteran Member)

Let me ask you this thing mate, what is the CTR% of the creative you are putting? I'm asking this since if you have a winner in the creative it just might be that the landing page the user wounds upon is simply not too engaging. Or it might be that there is a type of discrepancy between what the user is seeing on the push and from the landing page. Your promising him a hot girl and then the expectations don't meet reality, so the user doesn't want to carry on with the flow. in the end push traffic is all about dealing with real human beings.

The fact that your'e getting some conversion is nice though, maybe try more traffic sources, mix it up a bit and see if your can reach a positive ROI with a different mix of traffic with same offer and the same creative.

Anyhow mate, good luck


04-16-2019 10:05 AM #24 matuloo (Legendary Moderator)

Quote Originally Posted by mstgeorge View Post
And once again i was paused from the FR campaign for quality issues.
It is really frustrating having this happen once more so i took the decision to stop running dating with push.
I know a lot of people are running dating with push and making tons of money but for me doesn't work as i always have problems with lead quality.
So i have a decision to make here and i wanted to ask the opinion of more experienced people here.
Should i switch traffic source?(run adult dating with banners)
Or switch vertical?(run sweeps with push)
Push is like that unfortunately, at least with dating. The offer either can or can't monetize these leads properly. I've lost multiple offers this way, but then again, some even bumped my payouts. I know it's frustrating, but that's how it is with dating on push.

I'm also running dating with banner traffic and the quality problems are way less frequent there, they do occur of course, but not as often as with push traffic. And the same LPs that work with push can also work with banners, so it's definitely an option.


04-16-2019 11:34 AM #25 mstgeorge (Member)

I want to thank everyone that helped me in this follow-along especially matuloo and maynzie.
Despite the result i have learned so much on so little time and i hope other members have benefited too.
I think the next step that makes more sense for me is move to banners so unfortunately for the experienced members lol i will start another follow-along there to seek their wisdom


04-16-2019 12:19 PM #26 matuloo (Legendary Moderator)

Quote Originally Posted by mstgeorge View Post
I want to thank everyone that helped me in this follow-along especially matuloo and maynzie.
Despite the result i have learned so much on so little time and i hope other members have benefited too.
I think the next step that makes more sense for me is move to banners so unfortunately for the experienced members lol i will start another follow-along there to seek their wisdom
You're welcome And feel free to start another thread, will try to help you again


04-17-2019 02:40 PM #27 AdMaven (Veteran Member)

Quote Originally Posted by mstgeorge View Post
I want to thank everyone that helped me in this follow-along especially matuloo and maynzie.
Despite the result i have learned so much on so little time and i hope other members have benefited too.
I think the next step that makes more sense for me is move to banners so unfortunately for the experienced members lol i will start another follow-along there to seek their wisdom
Hey man, I'm not sure banner would bring you better results. In the time iv'e been in the Ad:Tech industry i noticed one thing about banner in the last year, banners are all over the place, There is so much of them users have what's called "Banner Blindness" - they just ignore it and carry on. Think about it, you go in a site with a distinct purpose to do something specific, the banner your seeing youv'e seen it in at least 100 different variants everywher, but when your using push your getting to more users in a much higher rate then with banner. If your going with dating i would test some more, try more sources, ask your AM what works, don't get dishearted on this ad format, I'm biased, but it's only because i see great results with this ad format with dating


04-17-2019 08:19 PM #28 matuloo (Legendary Moderator)

Quote Originally Posted by am2015 View Post
Hey man, I'm not sure banner would bring you better results. In the time iv'e been in the Ad:Tech industry i noticed one thing about banner in the last year, banners are all over the place, There is so much of them users have what's called "Banner Blindness" - they just ignore it and carry on. Think about it, you go in a site with a distinct purpose to do something specific, the banner your seeing youv'e seen it in at least 100 different variants everywher, but when your using push your getting to more users in a much higher rate then with banner. If your going with dating i would test some more, try more sources, ask your AM what works, don't get dishearted on this ad format, I'm biased, but it's only because i see great results with this ad format with dating
Banner do better than push, when talking about dating and lead quality, it's not just my opinion, I got it confirmed from several affiliate networks too. Bannerblindness is a thing for sure and it's causing banners losing volumes year after year, but when talking about lead quality alone, it's not really relevant. We are talking about people who already clicked the banner, not those who ignored it

I'm not saying push isn't working for dating either, many people are working with this format and getting solid results. It just required the right offers.


Home > Push Traffic >