Home > >

---

Newbie Follow Along with Push Traffic - DE Dating (31)

---

Hi All,

I'm starting this thread today with my journey of running offers in Push traffic. Special thanks to twinaxe who started the Push tutorial which gave me courage to try push.

A little background of me: I started Vortex's 40-day tutorial in Nov last year and finished it in Feb 2020. Found my first winning offer running Pop in May and make a few hundred bucks. Since then, I kept losing money. What's worse, I got fired from my formal job 10 days ago. Affiliate marketing is the only thing I have now.

What I want to achieve: hope I can get consistent XX per day after this thread ends. When I started the next follow along, the goal would be XXX per day

Day 1:
Offers: 3 dating offers from 2 networks. One offer has 2 offer links, so that adds up to 4 offers in my tracker indeed.(I chose dating offers because dating usually converts well but easy to encounter the quality issue. Still I started from dating to give myself a bit confidence)

Geo: DE

Payout: $2

Conversion point: SOI

Traffic source: PropellerAds. I chose Propeller because they usually approve campaign fast.

Landers: Ripped 5 landers and fixed a bit based on the Propeller policy.

Creatives: Selected 1 image icon with a woman and added "1" in the icon, just like this:


I didn't use Message icon like twinaxe mentioned in the guide in the first test. Because I thought it was dating offer which might perform better with a lady icon.

Titles and Descriptions: a total of 4 with emojis.

Launched the campaign with CPC of 0.038USD and selected users with high activity only.

Propeller suggested to split the campaign into in-page push and classic push. So this campaign will become 2 actually.

Let's see the result soon.

---

Just as I launched the campaign for 30 min, I discovered in-page push campaign was spending money fast. After checking the campaign itself, it looks like in-page push need lower bid to get enough traffic to test.


So I immediately lowered the bid to $0.016.

Really have no idea how much bid I shall set. Just checked the chart and selected the bid which might bring 5K+ clicks to the campaign. Hopefully, it will bring enough clicks to test the title. Otherwise, I can only copy the campaign again to test.

---

Hey, nice to see your FA

I wish you good luck with new traffic type, keep us updated.

One small thing - something is wrong with the linked images - did you embed it using img tags?


Sent from my iPhone using STM Forums mobile app

---

Day 2:
After running the campaign about 1 day, this is what I can see in Propeller dashboard:


CTR might looks okay, but CVR looks very low.

If we check the landers stats, the overall CTR is just 6.33% which might be very low. But if we look at performances of each lander, lander 4 looks promising. It's already green. Shall I just keep testing with lander 4 and pause the rest landers or just pause the worst lander?


Among the 4 offers in the tracker, 2 offers have 0 conversions, the other 2 have 8 and 10 conversions respectively. So, pausing the first 2 offers without conversion is a good choice.

If we look at the title and descriptions combined the 2 campaign stats, Ad copy No.1 has the highest CTR and CVR is not bad. It's a winner.


So, in the following day, I'm gonna go with ad copy No.1 with 4 different icons. Will post the results then.

---

I inserted the images as twinaxe taught in this post. Just use the BB code from imgur.com

https://stmforum.com/forum/showthrea...-In-Your-Posts

The images looks fine from my side. Not sure what's the reason.

---

Yes, dating is a very visual vertical where girls can work better than message icons.

I always keep classic push and in-page separated.

They are too different to run them together.

I always test on classic push first when you want to run push traffic because in-page is rather a mix of push and pops and banners.

Use this landing page to test offers, didn´t reach real statistical significance yet but for me it would be good enough



Yes

I wouldn´t call it a winner yet.

What I like to do in situations where it´s not that clear is not to look for 1 winner but rather get rid of the worst elements step by step so that sooner or later you are left with only good converting ones.

In this case, kill ad copy 4 and keep the other running until you can kill the next one and so on.

---

Last post looks good but previous ones says that requested image does not exist.

Are those images ending with .png?



Sent from my iPhone using STM Forums mobile app

---

yep, as you suggested, I'll just pause the copy 4 and let the campaign continue to run.

Today the campaign didn't get enough traffic. Only generated 3 conversions. Looks like the Ad CTR is not high enough, so Propeller deliver less traffic to it.

In order to give it more traffic, I can only pause the old campaign and duplicated it, increase the bid to $0.04 and launched the duplicated campaign(Classic Push only). Let's see if we can get more clicks this time.

---

Yes, they do.

Seems they are deleted on imgur

---

Day 4:

If we look at the performances of Ad copy in the past day, the copy 1 wins again. Not sure if the stats are enough:


Ad copy 1 have both higher CTR and CVR. Shall I continue testing it, give it more stats or mark it as a winner and starts icon test?

---

Although it didn´t reach real statistical significance yet I would continue with copy 1.

Copy 2 doesn´t look good and copy 3 would need to convert alot better than now to catch up with copy 1 and this would be very unlikely.

---

Thanks twinaxe for the follow up. btw, what's the tool you're using above to show the chart and graph. It look so cool.

Day 5

I took the Ad copy 1 with 4 different style of icons. The purpose of this is to test which type of icon get higher CTR and CVR.

After running the campaign for 1 day, this is what I got. Clearly Icon 1 & 2 Stand out. But icon 1 has higher CTR but a bit lower CVR whereas icon 2 has lower CTR but higher CVR. Hard to say which is the winner. Probably test icon 1 and icon 2 in separate campaigns to see the difference.

---

https://marketing.dynamicyield.com/bayesian-calculator/

---

Thanks Plutus.

It's very useful.

---

Day 6

Today I set up 2 campaigns to test icon 1 & 2 to see which one is better. However, the results are even worse than the former campaign. From the stats, Icon 1 seems to be a bit better but it's not a winner yet.


I felt so puzzled at this point. Propeller doesn't give enough traffic, making the whole testing process so slow.

This is what I though: I'd take either of the above icon and started testing images.

---

One tip, don´t always look for a winner.

Often it´s helpful to rather do it other way around and look for a loser or losers.

Then you don´t have the pressure to find the one winner, when you just try to get rid of the worst elements step by step it can also help to make faster decisions.

What do you try to get more traffic?

Best way is to create a new campaign everytime you change something.

New creatives?
- Make a new campaign

Getting rid of bad creatives?
- Make a new campaign and continue with the better ones

---

I looked at your Push tutorial in creative part, it looks like you're teaching us to find winning creatives. It urges me always look for the winner.

Now it looks clear. We need to find winners by abandoning the worst ones gradually. Is my understanding correct?

Thanks.

---

Well, winning creatives yes and the ultimate goal is to find a winner but I prefer to achive that goal step by step.

When you only look for that one winner you risk to keep bad elements running for too long = wasting too much money on it.

So instead of waiting until one element can be seen as a clear winner I mostly prefer to keep all elements running that could be a winner and rather eliminate the ones that have no chance to win.

The end result is the same but doing so can save tiem and money

---

Hey twinaxe,

This was my understanding after reading the push toturial:
1.Test Title & description:run multiple tests to find out the best performing title & description.

2.Test icon, Use the above best title & description to run multiple tests to find out the best performing icon;

3. Test big image: Use the winning icon, title&description in above steps to test big image.

But after reading your reply here:

I think this might be what you actually means in the Push tutorial:

Test title&description, icon & big images in separate campaigns.

Run these campaigns at the same time and cut the worst elements during the testing.

Find out the best performing elements(title&description, icon, big images) from testing campaigns and make new combinations from winning elements. Use the winning elements to continue test Landing pages, etc.

Is my new understanding correct?

---

I'm also a fan of this approach I used to call it the "cut the worst method" and I also wrote a short article about it quite a while ago, check it out: https://stmforum.com/forum/showthrea...-of-my-methods

---

Unfortunately, I just got the feedback from Affiliate manager that advertiser asked for pausing my campaign because of bad quality.

Dating offers looks like easy to be involve in quality issues. I may try some sweeps offers later.

My goal of achieving XX per day profit can't be realize at this thread. Anyway, thanks all of you for giving me the suggestions.

I'll start a new thread when I can see the performance of new campaigns.

See you in the next thread.

---

@fuyuzhelianmeng Let me try to clarify a bit

The first and most important step is to find a good offer.

Without a good offer all other tests are much harder or even impossible to run properly.

And to find a good offer you don´t need a perfect and highly optimized funnel right away.

Basically you only need a funnel that´s good enough to drive targeted and good quality traffic to the offer.

To do so I run the first test like this (focus is on how to test creatives):

- Creatives without big image for faster testing
- Few different title and description combinations

Often it happens that one creative then receives most of the volume and the others cripple more or less with very low volume.

Then I check my stats, when the creatives with low volume also don´t convert good I just focus on the creative that brings traffic.

When a creative with low volume converts good I test it in a separate campaign.

When I then have a good LP/offer combination I start more tests on creatives.

These tests are stuff like

- targeted/buy intended text vs broad/more clickbait like text
- message icon vs offer related icon
- big images
- with big image vs without big image

For each test I run a separated campaign.

The results are noted in an excel table like this


My original tables are a bit different but probably impossible to understand for someone else

With such tables I have an overview about all different tests and can find the best combinations easier.

Maybe there are better or other approaches for it but for me this works pretty good.

And yes, I run all these campaigns at the same time

Unluckily this is pretty common for dating, the advertisers there are more strict when it comes to quality.

---

I wouldn't run away from dating just because you lost 1 offer... it was your first lost offer, right?

It's very common to get kicked from offers due to quality in pretty much any vertical. I'm not sure whether sweeps are more lenient when it comes to quality, but I doubt there is such a big difference. As long as you keep on promoting lead based offers (SOIs, DOIs...) you will simply have to get used to this. I'm loosing offers all the time as well, it's absolutely normal.

On top of that, offers that are bad at monetization at the backend are not suitable for massive scaling anyways, you need to find advertisers who really know what they are doing and do have everything setup properly. Those with more robust monetization methods can also accept a bit lower quality leads. It's all about testing as many offers as possible.

---

Yep, this was the first offer I got kicked out. Felt a bit frustrated but still there're loads of offers there.

When I started running DE dating offer, I have tested 8 offers and this one converts very well and looks very promising.

Another reason for me to skip it is bid and volume in DE in propeller. DE seems to be a very competitive geo. I usually bid 0.056~0.06 CPC but can only get 600~1000 clicks per day.

As you suggested, I may look for some more offers but before I have suitable dating offers, I'll try sweeps and start a new thread.

Thanks Matuloo for your kind advice.

---

DE is a bit tricky GEO for dating in general, personally I've been struggling with it for years as well... I can get offers profitable, but they never last too long. I'm doing better in other GEOs. Then again, I personally know some people who are doing well there... not everyone can make all the GEOs work, that's the way it is

You can also consider other traffic types, not sure if you promoted mainstream or adult dating... but there are many solid adult traffic networks out there and dating happens to be the most promoted vertical. There are always many options to explore, don't let one failure discourage yourself!

---

Hello, try to take offers directly from advertisers, here are some solid ones on DE : Trafficpartner & Rebll

---

Yup, both great companies to work with, running for a while and reliable.

---

Sure. I've been getting lots of red campaigns in the last 7 months. But yeah, I'm still trying.

I think that offer is adult dating. There're lots of naked ladies in the offer page.

---

Ok, sounds like an adult offer for sure then. It's very common to run into quality problems with adult dating on push traffic. It's better to use adult sources for this, at least based on my experience. But anyways, looks like you decided to switch verticals so I won't give you more advice here. Should you decide to give it another go, lemme know

---

Thanks Matuloo

---

Trafficpartner is great, was running their offers when they were Sexmoney in 2008 or so already.

One tip though, usually they only offer Rev Share when you start and only offer SOI/DOI when you show solid performance already but Rev Share can be (much) harder to run compared to SOI/DOI, especially in the beginning.

Their offers are also available on networks like Clickdealer as SOI/DOI so it can help to first run them as SOI/DOI on a CPA network to work on the funnel and when you have a good converting funnel use it to run the offers as rev share.

This way you can avoid too heavy losses in the beginning and still get the longterm revenue from rev share.

And some offers can bring in rev share profits for very long time, even for years.

---

Home > >