Follow Along cyrus791's Journey Using Bemob + Mobipium + Popads + Luke's Pops Guide + Evadav

[cyrus791]

**July 16**
Affiliate network: Mobipium
Here's my result by targeting Argentina and WebsiteIDs:

It is going negative day by day. I am about to throw the towel on this campaign because I already narrowed down the targeting and I don't see any signs of green. Thoughts?

Affiliate network: Evadav
Here's my result after targeting Indonesia, Android, Chrome, and WebsiteID's:

Also, it is already narrowed down and I don't see any signs of green or breaking even. I really need help with this

On the side, I am also running a campaign targeting only in Indonesia to buy more data:
Here's the result:

I am constantly gathering converting WebsiteID data in this "Global - PopAds - test" campaign and I put it on "Indonesia - PopAds - Android_Chrome_ID_WebsiteID" campaign. Is that the proper thing to do to see some greens?

I really need some help.

 

[Jimmy]

Hi @cyrus791 - I've got a couple of questions so I can try to help you out...

I duplicated WW_Android_Chrome_Argentina_WebsiteID and created WW_Android_Chrome_Argentina_WebsiteID1.

What do you mean by duplicate? What is the difference between WebsiteID and WebsiteID1?

I am constantly gathering converting WebsiteID data

Could you explain a bit more detail about what your process is for WebsiteID data? I assume you mean that you're creating campaigns with a publisher whitelist on Popads, but I'd like to find out what criteria you're choosing for selecting a WebsiteID? Are you just choosing publishers with conversions or is profitability also a factor and if so, what kind of statistics are you using to choose the winners?

 

[cyrus791]

Hi @cyrus791

What do you mean by duplicate? What is the difference between WebsiteID and WebsiteID1?

They are basically the exact same targeting, I just put "1" to the "WebsiteID1" to make it different in my tracker.

Hi @cyrus791
Could you explain a bit more detail about what your process is for WebsiteID data? I assume you mean that you're creating campaigns with a publisher whitelist on Popads, but I'd like to find out what criteria you're choosing for selecting a WebsiteID? Are you just choosing publishers with conversions or is profitability also a factor and if so, what kind of statistics are you using to choose the winners?

For example, I got a conversion on WebsiteID 4020143 from my campaign "Global - PopAds - test", I put 4020143 on a notepad. The 4020143 WebsiteID will be used on a new campaign named "Android_Chrome_ID_WebsiteID".

It was named "Android_Chrome_ID_WebsiteID" because I included the WebsiteID's on this new campaign. I gathered these WebsiteID's from "Global - PopAds - test".

The criteria I used in selecting WebsiteID's is when I see positive ROI, I take note of them.

Hi @cyrus791
Are you just choosing publishers with conversions or is profitability also a factor and if so, what kind of statistics are you using to choose the winners?

Profitability in these WebsiteID's is a factor that I choose. Everytime I see ROI on WebsiteID's I include them.

I hope you understand my point. I tried my best to explain.

 

[Jimmy]

basically the exact same targeting

by 'basically' the same, do you really mean exactly the same or are there some differences? The reason I ask is that if you're running two campaigns that are exactly the same then you're bidding against yourself and doing double the work by having to manage two campaigns instead of one. What was the reason you decided to duplicate?

It was named "Android_Chrome_ID_WebsiteID" because I included the WebsiteID's on this new campaign. I gathered these WebsiteID's from "Global - PopAds - test".

The criteria I used in selecting WebsiteID's is when I see positive ROI, I take note of them.

OK, so just to check I'm understanding; you're running two campaigns:

1. "Global" with a blacklist - i.e. you're targeting all publishers except the ones you remove to the WebsiteID campaign
2. "WebsiteID" with a whitelist - i.e. you're only targeting publishers with a positive ROI from the Global campaign

How are you optimizing the WebsiteID campaign? I mean, what kind of metrics are you looking at to decide whether to remove a publisher from this campaign (e.g. if ROI drops below x% after $x spend or something like that)?

 

[cyrus791]

by 'basically' the same, do you really mean exactly the same or are there some differences? The reason I ask is that if you're running two campaigns that are exactly the same then you're bidding against yourself and doing double the work by having to manage two campaigns instead of one. What was the reason you decided to duplicate?

I mean they are exactly the same. The only difference is the name of the campaign itself.

OK, so just to check I'm understanding; you're running two campaigns:

1. "Global" with a blacklist - i.e. you're targeting all publishers except the ones you remove to the WebsiteID campaign
2. "WebsiteID" with a whitelist - i.e. you're only targeting publishers with a positive ROI from the Global campaign

How are you optimizing the WebsiteID campaign? I mean, what kind of metrics are you looking at to decide whether to remove a publisher from this campaign (e.g. if ROI drops below x% after $x spend or something like that)?

We probably have a different level of understanding when explaining things. We are not on the same page.

I am probably mixing it with Mobipium and Evadav campaigns. That's why it creates confusion. I apologize.

Let me explain one more time until we are on the same page.

Please ignore all the recent posts for now. Let's start over from scratch with these campaigns:

These campaigns are using Mobipium smartlinks.
The first and second campaign are being setup like this on PopAds.

Next is this campaign:

This campaign is running Mobium Smartlink too. This is where I extract positive ROI WebsiteID's in Argentina.

Here's the targeting of "WW_Android_Chrome" campaign in PopAds:

Now, let's go to Evadav offer campaigns.

Here's the targeting for "Android_Chrome_ID_WebsiteID"(Android,Chrome,Indonesia,WebsiteID)

Just in case you are wondering what is the ID after Chrome. It is Indonesia. Don't be confused with "ID" country and WebsiteID.


Next is this campaign:

This campaign runs Evadav push notification offers.

This is where I extract positive ROI WebsiteID's in Indonesia.

Here's the targeting in PopAds:

I hope I made myself clear this time about the targeting.

I tried my very best to explain my current situation.

@jimmyvanilla, I hope we get on the same page. Thank you for helping bro.

 

[Jimmy]

I hope I made myself clear this time about the targeting.

I tried my very best to explain my current situation.

@jimmyvanilla, I hope we get on the same page. Thank you for helping bro.

This is great. I know it was a bit of a task to include those screenshots, but I see what you're doing now. I've had a few ideas, but wanted to make sure I understood the situation.

I'll specifically talk about your campaign for Argentina for MOBIPIUM, but the principle is the same for the Indonesia campaign for Evadav. The thing that jumps out at me straight away is that you're targeting the same publishers (i.e. WebsiteIDs) in 3x campaigns (2x campaigns for Indonesia), so you're competing against yourself in terms of buying traffic from those publishers.

Now I'm not really sure how much background to give here so bear with me if I'm getting too basic with any explanations of how the traffic bidding process works, but essentially, what happens whenever someone visits a website with Popads code is that a popunder ad appears and all the advertisers bidding for that ad (i.e. with targeting that matches the visitor) are thrown together in a kind of auction where (essentially) the highest bid wins. If a visitor from Argentina comes along, all three of your campaigns are trying to show an ad to that visitor, but only one can win. Then when the next visitor comes along the same thing happens and one wins, etc.

This means you end up getting only a small part of the "data picture" painted on each campaign:

Whereas what you need to be able to optimize effectively is to see all the data together in the one campaign:

If you want to have one Argentina campaign with a whitelist, what you should do is:

1. Stop both WebsiteID and WebsiteID1 campaigns
2. Go back to your data from your Global campaign and find all the publishers for Argentina that:
   1. Have the ROI you want to target
   2. Have terrible ROI that you're happy to cut forever
3. Start a new WebsiteID2 campaign targeting Argentina and include (whitelist) the good websites
4. Start another Argentina campaign WebsiteID3 and exclude (blacklist):
   1. The websites you're including in your WebsiteID2
   2. All the really bad websites from your Argentina data from your Global campaign
5. Exclude Argentina from your Global campaign going forward

With this approach, what you're going to do in the future is:

1. Remove websites that don't work out from your WebsiteID2 campaign
2. Add the good websites from WebsiteID3 into Website ID2
3. Exclude any more bad websites from WebsiteID3

Alternatively, if you're happy to keep testing one Argentina campaign, what you could do is:

1. Stop both WebsiteID and WebsiteID1 campaigns
2. Go back to your data from your Global campaign and find all the publishers for Argentina that:
   1. Have terrible ROI that you're happy to cut forever
3. Start a new WebsiteID2 campaign targeting Argentina and exclude (blacklist) the bad websites
4. Exclude Argentina from your Global campaign going forward

With this approach, what you're going to do in the future is:

1. Keep excluding (blacklisting) websites that don't work out from your WebsiteID2 campaign

The second approach is simpler, but either approach will help you optimize properly.

 

[cyrus791]

This is great. I know it was a bit of a task to include those screenshots, but I see what you're doing now. I've had a few ideas, but wanted to make sure I understood the situation.

I'll specifically talk about your campaign for Argentina for MOBIPIUM, but the principle is the same for the Indonesia campaign for Evadav. The thing that jumps out at me straight away is that you're targeting the same publishers (i.e. WebsiteIDs) in 3x campaigns (2x campaigns for Indonesia), so you're competing against yourself in terms of buying traffic from those publishers.

Now I'm not really sure how much background to give here so bear with me if I'm getting too basic with any explanations of how the traffic bidding process works, but essentially, what happens whenever someone visits a website with Popads code is that a popunder ad appears and all the advertisers bidding for that ad (i.e. with targeting that matches the visitor) are thrown together in a kind of auction where (essentially) the highest bid wins. If a visitor from Argentina comes along, all three of your campaigns are trying to show an ad to that visitor, but only one can win. Then when the next visitor comes along the same thing happens and one wins, etc.

This means you end up getting only a small part of the "data picture" painted on each campaign:

Whereas what you need to be able to optimize effectively is to see all the data together in the one campaign:

If you want to have one Argentina campaign with a whitelist, what you should do is:

1. Stop both WebsiteID and WebsiteID1 campaigns
2. Go back to your data from your Global campaign and find all the publishers for Argentina that:
   1. Have the ROI you want to target
   2. Have terrible ROI that you're happy to cut forever
3. Start a new WebsiteID2 campaign targeting Argentina and include (whitelist) the good websites
4. Start another Argentina campaign WebsiteID3 and exclude (blacklist):
   1. The websites you're including in your WebsiteID2
   2. All the really bad websites from your Argentina data from your Global campaign
5. Exclude Argentina from your Global campaign going forward

With this approach, what you're going to do in the future is:

1. Remove websites that don't work out from your WebsiteID2 campaign
2. Add the good websites from WebsiteID3 into Website ID2
3. Exclude any more bad websites from WebsiteID3

Alternatively, if you're happy to keep testing one Argentina campaign, what you could do is:

1. Stop both WebsiteID and WebsiteID1 campaigns
2. Go back to your data from your Global campaign and find all the publishers for Argentina that:
   1. Have terrible ROI that you're happy to cut forever
3. Start a new WebsiteID2 campaign targeting Argentina and exclude (blacklist) the bad websites
4. Exclude Argentina from your Global campaign going forward

With this approach, what you're going to do in the future is:

1. Keep excluding (blacklisting) websites that don't work out from your WebsiteID2 campaign

The second approach is simpler, but either approach will help you optimize properly.

That's a loooong but informative read. I will apply this principles in Evadav push notification offers. My test there looks promising. I might have a winner there. I am thinking that it will take longer for Mobipium campaigns to go green because of poor payout but I am seeing Evadav push notification fixed payouts will give me faster green results.

 

[Jimmy]

My test there looks promising. I might have a winner there.

Sounds cool! Keep us posted.

 

[Akahma]

Now I'm not really sure how much background to give here so bear with me if I'm getting too basic with any explanations of how the traffic bidding process works, but essentially, what happens whenever someone visits a website with Popads code is that a popunder ad appears and all the advertisers bidding for that ad (i.e. with targeting that matches the visitor) are thrown together in a kind of auction where (essentially) the highest bid wins. If a visitor from Argentina comes along, all three of your campaigns are trying to show an ad to that visitor, but only one can win. Then when the next visitor comes along the same thing happens and one wins, etc.

Great info @jimmyvanilla.
i have one comment to add regarding multiple campaigns, and I'm doing this Especially with Evadav
For +ROI GEOS
Instead of increasing frequency capping for any campaign especially when testing multi landers, I'm creating a multi campaign with the same frequency capping that is 1 every 24 hours, with a little /even very little variation on bidding,
here I'm not competing with my self, here I'm getting all the traffic from this source and test it.
you can check it in your live bidding chart on popads as sabotaged
and I even got the same impression for the second campaign lower,
for example, I'm targeting Italy and my main campaign cpm is .0005 and in the 2nd I'm making it .00045 the third .0004 .
all with the same targeting, put in mind there is a click loss, and its Vary a lot with each GEO especially with evadav , so according to the click loss also i do duplicate campaign instead for increasing frequency capping

 

[Jimmy]

i have one comment to add regarding multiple campaigns

Interesting. Do you have a specific reason for doing this rather than increasing your frequency capping on a single campaign?

 

[Akahma]

if im testing more than one offer/ landers, Also if I'm Targeting "exclusively and need all impressions of this site exactly"

 

[Akahma]

Whereas what you need to be able to optimize effectively is to see all the data together in the one campaign:

to be more clearer... For site ID 970237 that site id is promising for Streaming

it has the considerable quality and only 14 campaigns are competing with me and I have my targeting and also I have one step over,
I know the domain name " I have investigated it before early on anther campaign and its already on my list" https://www.estrenosdoramas.net
So if this site is profitable to me, I investigate the code for popads inside it to check its restrictions . and exclusively for this example, I 've found this site sending over than one pop for each user, and making a delay .( also could be known from the huge difference from raw traffic 389000 and unique traffic 98.145 . and according to this, I will make more than one offer and try to make the lp load as fast as possible ...
if impressions of this site are still good,I make the lp exclusive for this site and adjust it to the theme of the site,"will worth it " then make more than 2 or three campaigns for this site only .. grabbing as much as from it

 

[cyrus791]

Great info @jimmyvanilla.
i have one comment to add regarding multiple campaigns, and I'm doing this Especially with Evadav
For +ROI GEOS
Instead of increasing frequency capping for any campaign especially when testing multi landers, I'm creating a multi campaign with the same frequency capping that is 1 every 24 hours, with a little /even very little variation on bidding,
here I'm not competing with my self, here I'm getting all the traffic from this source and test it.
you can check it in your live bidding chart on popads as sabotaged
and I even got the same impression for the second campaign lower,
for example, I'm targeting Italy and my main campaign cpm is .0005 and in the 2nd I'm making it .00045 the third .0004 .
all with the same targeting, put in mind there is a click loss, and its Vary a lot with each GEO especially with evadav , so according to the click loss also i do duplicate campaign instead for increasing frequency capping

@akahma, thanks for the share about doing bid variations, I will give it a shot.

Here's what I am doing to win bids by increasing bid value by one decimal place. Please take a look :

As you can see in the first graph on the left and 1st bar, 0.000051-0.000149, I would bid 0.0001491 - see the change? I just added one at the end. With this bid, you can win most of the 1st bar in the 1st graph on the left plus you can outbid 0.000149 in the 2nd bar which is the minimal bid on the 2nd bar. This will save you money by bidding wisely.

This is helpful when you are testing out campaigns with a very little budget.

 

[cyrus791]

**July 17**

I decided to stop my Mobipium campaigns and proceed with Evadav campaigns.

Here's my results for "test" campaign - this is only targeting Indonesia(don't be confused with Global as I forgot to change it):

This is where I gather converting WebsiteID data and put it on "Android_Chrome_ID_WebsiteID" campaign side of PopAds.



Here's my results for "Android_Chrome_ID_WebsiteID"

It is overall 35% because there are some issues on Evadav that made me spent without getting conversions. When I sorted it by offers without including the offer with a horrible conversion, I get -23%. I can see that my ROI is still negative but I sense that it is improving day by day after adding new WebsiteIDs that I extracted from "test" campaign.

Here's a screenshot of -23% ROI.

These are my top performing landing pages on Evadav.

Play Youtube 2
Play Video 2
Video 8
aa Video 3
Bot Check 6

Feel free to test it.

That's all for now.

 

[Akahma]

Great improvement, keep going
One last thing , with tier 3 countries you may face A lot of clickloss, keep tracking the amount of impressions you got from popads to tracker
Then check the amount reached to evadav , to do that you need also to adjust the time zone for all three(,popads, tracker, &evadav)

 

[cyrus791]

**July 18**
Here's my results for "test" campaign which targets Indonesia only:

I can see that it is improving because last July 17th, it was -33.35%

What I did to improve it was, I changed from these:

to only this:

Yes, I used only one offer page from Evadav.

Now, here's my results for "Android_Chrome_ID_WebsiteID " campaign:

I am almost breaking even but guess what? I saw green in more volume! I can see the potential but it needs more data to increase ROI.

I did the same thing to this campaign to improve ROI.

I changed from these:
Play Video 2
Play Youtube
Video 8
Video 7
Play Youtube 2

To this:
Play Video 2

In simple words, "Play Video 2" landing page is the best converting push subscription offer from Evadav for me.

See the impact of the change:

I changed it on 12:00-12:59.

That's all for now.

 

[Jimmy]

I can see that it is improving because last July 17th, it was -33.35%

Nice work. Keep it going!

In simple words, "Play Video 2" landing page is the best converting push subscription offer from Evadav for me.

Did you use any kind of calculation or tool to make this decision? If you're interested in reading more about statistical significance, I wrote a detailed guide on statistics and split testing here.

Also, the data you posted didn't seem to indicate that Play Video 2 was the best, but I know you've probably got other data you're looking at.

 

[cyrus791]

Did you use any kind of calculation or tool to make this decision? If you're interested in reading more about statistical significance, I wrote a detailed guide on statistics and split testing here.

No. I just based it on this data:

I know its not a lot of data because of 48 conversions for Play Video 2 and the difference from the 2nd rank is 8 conversions. That's the basis of my decision to use it and it was effective for me.

If you're interested in reading more about statistical significance, I wrote a detailed guide on statistics and split testing here.

I will have a look. Thanks.

 

[Jimmy]

I know its not a lot of data because of 48 conversions for Play Video 2 and the difference from the 2nd rank is 8 conversions. That's the basis of my decision to use it and it was effective for me.

When it's working, work it!!

 

[cyrus791]

**July 19**
Here's my results for "test" campaign:

This result is much worse compared to July 18th result. The ROI last July 18 is -20.51% and this July 19 it was -52.51%. I didn't expect to get better because I am targeting all websites in Indonesia.

Now, here's my results for "Android_Chrome_ID_WebsiteID " campaign:

I saw green day! It is improving.

What I did to get to this point is, I extracted positive ROI WebsiteIDs from "test" campaign and put it in this campaign.

I do that every day.