Home > Mobile > Follow-along Campaigns

---

Never surrender! (35)

---

1. Stop creating banners and landers yourself
2. Get yourself an adplexity subscription and spend 80% of your time researching / reengineering what the guys, who are running scale, are doing. In the rank of importance GEO>Offer>Traffic Source>Creatives
3. Rip their landers and creatives 1:1
4. Only test offers that you see others running, or are on the networks top list, or are recommendation by your AM.
5. If an camp shows no initial traction kill it and move on. If it shows traction, improve upon it.
6. Spend more. With 10$/day budget you won't find an winning campaign till 2018
7. Not that important, since you should kill the camp anyway, but the lander speed sucks. 1.4mb size is not good for mobile. Speed is the name of the game. Go for 200-300kb size.
https://gtmetrix.com/reports/www.the...g.com/9HqeksZ8

Edit: For sweepstakes, "chance to win" and no social proof will hardly fly on display. Need to be more aggressive. But adplexity will show you the way.

---

Hi Sebastian, thanks for your reply, it's precious. Just a few questions

Are you saying this because I may not have the knowledge to create them in the proper way or because ripping creatives is the best strategy?

Don't you think that AM may recommend offers that are new or that many other people are running? In this way I may end up promoting something that's saturated or playing the "guinea pig" part for the AN?

That's what I did

Oops. I made a mistake while writing. I usually set my daily budget at 20/30$. Is it enough?

Correct! My problem here are the image. It tends to make the page heavier. That's another reason why I removed it in the second LP.

Thank you!

---

Winning creatives evolve out of hundreds or thousands initial tests and improvements of already winning creatives. Beating the control group (the current creatives which are working) is even tough for experienced guys. For a newbie, its like winning the lottery

With your AM it depends on how you ask. I always ask for (i) offers that are currently doing solid damage (ii) new offers that have already shown initial traction. After a couple weeks or months your AM, if he's a good one, gets a feeling for what you can make work with your traffic and will give you proper recommendations. But thats only 30% of the work. Live spying and spying on adplexity for offers is as important.

When I started out, I was spending 50-100$/day. That summed up to about 2k initial loss, of course I made some revenue during that period, and found my first almost 1k/day camp within 2-3 months.

At the end of the day it comes down to money vs. time. IF yoou spend little, it will take a long time to find a winning offer and to get meaningful data. If you spend more, it "can" get quicker. You "only" need to survive long enough (not running out of budget). But money you can always get back, time not.

---

Sebastian is right, thou I'm all for creativity, when you are new, the best thing you can do is to copy those who already have experience.

You don't have to be a 100% copycat - tweak the creatives, play with colors, change imagery, change wording a bit ... but do not try to come up with your very own creatives from scratch. Not until you understand what makes a good LP.

$20-$30 per day can be enough if you limit your targeting - do not target everyone but make it more specific based on what your current traffic source supports, for example : Android, Wifi, smartphones only ...

Just one more tip : If you decide to continue with banners, you need to reach way higher CTR too - 0.3% is not enough for such GEOs. Looks for placements that are giving you impressions but no clicks and cut them, there will be some for sure.

---

Great suggestions, Sebastian. Thank you!

Correct. Regarding the screenshots I posted above, do you think I cut the campaign too early in order to collect those meaningful data? Did I have to let it run a bit longer?

I just love to make banners and landing page and have fun trying to make it fit in every screen but I admit that your advice mirrors reality . I also have the feeling that my banners don't catch the eye. I usually have 0,2-0,3% of CTR in every GEO and campaign I launch. Maybe with ripped banners it will go up.

Thanks for the last tip. I surely use it in the next campaign

---

Letting it run longer than $10 depends on the payout of the offer and how well the funnel of the campaign is developed, but I would say yes.

Yes in the following way:

When you start on a new network you get traffic from the worst placements you can imagine. Your friendly network sends you the traffic the big guys left there for the taking.

Further with display, you have many many moving components. Creatives, placements, lander, offer, bid & ctr, device targeting.

Each of those moving components can ruin even an superstar camp. You see, 10$ ain't too much to figure out so many components.

Therefore you have to remove as many moving components, or better said, borrow the results for those components from the guys who are already have done the testing. Use their banner, lander, device targeting, offer. So you are left with placements, bidding and establishing an high ctr. Still an tough job. Display can be brutal. The creatives and the importance of ctr makes the shit ten times harder than pops.

---

Well, If I have a payout of $0,70 and after almost $20 of spend I only have 1 conversion like in this case I suppose that stopping is the right thing to do, isn't it?

I see . I've realized it also by looking at the name of the placements in my traffic source. That's terrible. Could this problem be solved just by spending more in daily budget and bid?

It looks like you prefer pops . Sorry for newbie question but I have always known that spytools show you the creatives, how do I know the device targeting, Os etc?

Thanks for everything man

---

Yes, in case your funnel was at least decent and you didn't get all the traffic from a small amount of placements that happen to be bot-placements.

Yes and No. You need to learn how to detect the bad placements and block them. Then you can bid higher and spend more. If you don't block the bad placements prior to bidding higher, they will just send you more of that crap traffic.

You can only find out what the spytool collects - so adplexity for example will show you the ad type, iOS vs Android, wifi or carrier, size of ad, country, traffic source, affiliate network and tracking tool. And that's pretty much about it and to my knowledge adplexity is the best spytool right now.

---

In my tracker there are two columns called "bot visits" and "bot traffic" so I suppose that if they show "0" no bot traffic should be there.

Just to be sure, you said "if the funnel is decent" what do you exactly mean with that? Good banners-landers combinations?

Ok . The first time I used it there weren't all this tools . About Adplexity, I'm having a hard time with it maybe because I have the wrong expectation and how to get what I need out of it. I talked about it HERE. Could you help me please?

EDIT. Ok I spent a couple of hours on Adplexity and I'm getting ok with it .

---

Adding to this...

Not only that, but you should consider making blacklists for each vertical, and start new campaigns WITHOUT these placements. It's unlikely they will perform well in the near future.

In order to avoid very bad surprises, you could always just add them back for a test once a month and see what happens. Don't waste $14+ EVERY SINGLE TIME you start a new campaigns

---

Good point of course.

---

Thank you very much everybody!

These are the things I have learned this time and I was thinking while studying the data . I'm starting to believe that targeting the right placements is the key to victory.

I don't know if I should cut this campaign and move to another offer or try again. In my inexperience I'm not sure if this could become good enough. What would you do?

---

Go over the placements and cut everything with CTR below let's say 0.5%, then look at how much you would have spent on the placements that remained active. Now look how much you were actually paying per click. You need to make a realistic estimate - is there a chance that the traffic will convert good enough to cover the expenses? Check the winrates on the remaining spots too, are they usually higher than 30%? This means you can most likely go down with the bids too and get cheaper clicks.

I have no idea right now if you can make it profitable, but since you already have it all setup, go on with it for a while and learn from it. See how much you can get the CPC down by cutting the wrong placements. See if you can uncover some pockets of profit by doing this. Check where is the rock bottom for bids in order to still maintain good WR and CTR ... There is a lot you can learn from this campaign

---

DSPs got harder to optimize so it's not gonna be as easy as spending 10 bucks and getting profitable. In order to make them work now, you might have to do the following :

1. you need to analyze what bid levels give you access to the best placements and then whitelist them so you can bid high on just those
2. figure out what's the minimum bid to get decent WR on the lower quality placements that are still able to make conversions and again whitelist them as in point 1
3. in case of PIN submits, make sure you are only buying traffic from the carriers that the advertiser wants, you will need to use IP targeting a lot for this
4. constantly watch campaigns and cut placements with lousy CTR
5. you will need to test a ton of landers and offers and you need to either cloak or learn how to walk on the edge with your angles

It was way easier just a couple months ago, but that's not the case anymore.

In case you think this is too much of a hassle, I would recommend to switch to POPs for now.

---

Hi Matuloo,
First of all, thanks for your time

Of course, the higher you bid the better the placements are, and better placements mean higher revenue. Anyway I suppose it's all a matter of testing here. There's no way to know "how to bid", isn't it?

Ok, how do I know what IP ranges are the right ones?

This has been one of my sins. I've always tested no more than 3 landers for offer. From the next campaign on I'll use more landers

No it's not too much, I can do it and in any case I was thinking about this lately . I'm going to dedicate 1 day to mobile display and 1 day to mobile pops alternately every week. I'm in a personal situation where I need to make some money as soon as possible (of course always in the world of reality with AM) but I wouldn't want to waste all the time I've been spending studying display so I'll go on with mobile display but I'll start mobile pops too. If you think that this idea of mine could be useless or counterproductive please tell me

Next things I will do:
-Launch a new offer for mobile display
-Launch the very first campaign in mobile pops.

---

Right, testing is the only way to find out.

You can either get them from someone who has them (networks, friends, advertisers), some are available on the net (google is your friend) or analyze the data based on traffic you bought already and look for IPs that converted (most expensive way). There are also companies like digitalelement who have them, but not sure if they sell the data itself because its a on-request type of service.

Definitely focus on this more.

This could work. It's better to focus on one thing when starting out, but 2 should be doable too.

Keep us posted with your progress.

---

This is what Popads says about the quality:

Do not forget you can set this up also as a token in Voluum (token called [QUALITY]) and then you can collect data on this signal and later see if it's worth to optimise.
For start I would just choose "All quality". When I look in some of my campaigns I can see that even lower quality brings in nice profits and even higher ROIs then quality 10 for example.
So take this with caution.

On Popads you can sometimes get well with low bids. Rule of thumb is to bid a bit above average. But to properly test biddin gositions you can create campaigns at certain levels (HIGH BID, MID BID, LOW BID) and after decide which brings you the most profits.

---

Thank you Eric!

Your suggestion on quality was useful and I think it is a good point where to start from . About bidding I think I'll increase as needed

Do you have anu opinion/tip on my campaigns?

---

Hello again David!

Your follow along is another proof that mobile display got harder and POPs are once again the best pick for new guys to start with. I noticed you tried an adult offer, it might be a good idea to test some adult sources too - and not just POPs, try banners too, adult networks are different from the DSPs you are used to.

These adult offers for India and similar GEOs are usually carrier billing offers, so you need to buy just carrier traffic - did you do that? When running offers like this, it's better to use a tracker that has proper carrier detection, not sure how well Adsbridge works here as I didn't try this tracker myself yet.

What networks did you get these offers from btw, I found Kimia or Mobidea to work better for these compared to classic networks that offer a bit of everything.

---

Hello man!

Well, I don't know if it's hard or if it's just me that I'm not able yet . If you, or another expert affiliate, were me I'm sure you'd have made it work.
Can you make some name of this adult network? I found one called reporo.com

Yes I did it. In PopAds the carrier selection is available even if some "wi-fi guy" can get in.

Yes I used Mobidea .

These day I've tried another offer with pops : an AVG carrier billing. Spent 3,12 $ about 240 visits (impressions) on the pop, about 20 clicks on the CTA buttons but 0 conversions. I suppose this could be because I had to edit the landers which were too aggressive (so I had to change the "you have a virus" angle to the "you MAY have a virus" angle. Maybe this is what killed the campaign)

Do you have any suggestion to make this all display-pops mess go better?

---

The thing with DSP display ads is, it requires a lot to be blown on tests now. Just a couple months ago, you could get good traffic from the large apps or sites very cheap - these days, with the same bids, you will get 0% winrate or even 0 bids on them. So you gotta bid higher, then watch the bigger placements and put them into separate whitelist campaigns. It's way more complicated and you can't really even find these placements with small tests.

If you want to test carrier billing offers on adult traffic, exoclick is your best bet, they have he most volume so you should be able to find some good placements there.

You said you bought carrier traffic from popads - it's fine that they offer this targeting option, but in many cases, the carrier DBs of traffic sources do not reflect the reality. You need to double check this with a good tracker and in some cases, you need to target specific IP ranges - not all sources allow this tho - exoclick has it for example.

Test Kimia too, I send traffic to both, something works better in one, something in the other ... and there is also brokerbabe to consider.

---

One newbie question, why is it necessary to create a whitelist campaign on its own and not just keep those good placements in the running one?

Thank you very much for this informations . One doubt though. It looks like you don't recommend carrier billing offers (which could be considered CPS offers) on Popads. I wouldn't want to jump from one TS to another one again because I've already added PopAds to GO2mobi in my traffic sources "inventory" and I'm afraid another change/addition may be confusing and slowing right now .

In order not to make the same mistake I did with Go2mobi (go on creating campaign that are not suitable for that TS wasting time and money) can you suggest what kind of offers is more likely to get profitable on a traffic source like PopAds?

---

The reason is simple, different bid levels required to get traffic from different placements. It is not a problem on all sources, but with DSPs it is now. Say you are bidding $0.20 and you are getting traffic from many lower quality sources, your winrates are around 30% on those sources. Now you up the bid to $0.50, new and better quality sources start to send you traffic, which is good. But at the same time, you now pay higher CPM for the lower quality sources and your winrate on those is now 60-80%, which is something that you don't want. With every increase, the same happens to more and more placements. This is the problem, you need to segment the placements and target similar placements in separate campaigns.


This was not about popads specifically, I was talking more in general - in order to make carrier billing PINs work on a traffic source, it has to support carrier targeting. Popads has carrier targeting and they even support IP targeting I think, so you could definitely utilize this source for promoting PINs

---

Hello again davidep,

you don't have enough data to make a lot of optimizations, but indeed, there are some placements that look way worse than the rest. So in order to cut down the costs of testing, you can cut some of them - wait for 2x offer payout and if you don't see a conversion, cut it. 2x payout is not a lot, so you might want to retest some of these placements again later on, when you have a SOLID funnel narrowed down. At this point, it's crucial for you to make sure that you are buying only real traffic that is converting at least "somewhat". Then you can work on the LPs and find the best ones.

So, cut the worst one(s) and keep on running until you have enough data to cut some more.

I would also recommend to try to run for 24 hours, not just certain part of the day - at least try it to see if the performance goes up and down during the day.

Do not cut based on OS/device ... yet, you need way more data and a strong pattern to cut such a large part of your traffic.

---

Hi Matuloo!

I'd need to have some explanations:

What are you referring to exactly? I mean, I have tens of thousands of visits with more than 30x payout spend (offers payout is $0,50) I thought I had a lot of data . Or maybe you're talking about Os, device etc data? In that case yes it's true, all those elements have basically the same values

With real traffic you're probably talking about bot traffic, aren't you?
About LPs, I cut a couple in the beginning of the campaign because those were not promising at all.

I tested this in the past and morning-afternoon campaigns were a total flop for me. Doesn't that happen to you too?

One last question. By playing with the traffic estimator of Popads and by looking at the data it seems that a couple of placements has almost all the traffic. In fact if I keep only those placements the amount of traffic is still big. Maybe it is a stupid question but would it make sense to launch a campaign only on those ones?
If they're getting all the traffic and the most of the conversions that could be a good idea to be profitable?

---

Hi Davide!

It's nice to see that you are learning and growing every day!

Regarding your questions, I am curious for Matuloo's answer. Although from my side I'd say it makes sense to work with the top placements of course, but are those top ones matching the ones that are profitable for you? If so, you can create a kind of a white list to focus on those 2 and still try to work separately with the smaller ones in another campaign. I believe you have enough data to start optimizing.

I also checked the stats on Mobidea's side, and it is surprising the campaign went well with desktop, good! Be more careful with daypartying though, if you change the hours it is better to create a separate campaign, as if you mix, the existing campaign might behave very different, and mess your stats.

---

Here we go

At this point you have enough data to judge one thing only : whether the offer has promise or not. Since you got multiple conversions and your ROI isn't a -90% catastrophe, I would say it has some potential. For anything else, there isn't enough data, as it's spread across to many targets. You started by rotating 6 LPs and 2 offers, try to calculate how many combinations that is You need to spend more than $20 to be able to optimize a bit deeper.

Yup, in other words, getting rid of the placements that just eat up your budget and dont bring anything in. As for the LPs, careful with what you cut early too, at this point I would only cut those that have really lousy CTR.

I pretty much don't use day-parting now, I did in the past and I certainly will in the future again when needed but right now I try to make campaigns that can run 24/7.

This is the same with any traffic source, let's say 10% of placements are responsible for 50% of traffic - these are the large placements. Then there are the small ones that account for maybe 50% of the placements but only bring in about 10% of traffic. I made those numbers up, but you get the idea. Technically, you can target just those, that's the point of whitelist campaigns, but again, your goal should be to run on as many placements as possible to get maximum volume.

It's certainly something you can try thou, the logic behind it is valid. In case the largest placements are also converting the best for you, it's a good idea to focus more on those.

---

Campaign Update


The campaign is dying so I decided to stop it

While asking here for tips the campaign was going.

I let it run on Saturday during the afternoon and evening until midnight and it showed some convertions in the evening (with an awful afternoon performance). This weekend's data are basically the same as those I posted above: there is never much difference between the various elements and as usual only placements show difference. On Sunday evening it kept on having conversions only on the usual 4/5 placements but still in red because the low payout ($0,5) couldn't cover the cost of the traffic (0,8 CPM). Yesterday I gave the last chance to the campaign, run only with the best placements, but gave only 1 conversion.

Here are the final results (spend should be more than $40 instead of $34)


Oh BTW, I was going to use the staggered bid tactic by Vortex (3 campaigns: each with a different bid level) but because of the very big amount of traffic I received with a low bid it didn't seem to me the case to do it


@Mobidea: As for placements. Yes I think that the key point is to find the placements that are profitable, the point is that those could be the ones that don't receive a lot of traffic

I get what you mean. In this case, since all the elements but placements tend to be evenly matched, I thought that focusing only on those was a good idea. Maybe next time, I should keep on until all the others elements show difference. Correct?

I understand. As I told Mobidea, it's a matter of finding the profitable placements and they don't have to be necessarily those that receive the most of the traffic. Correct?
And of course I included as many placements as possible to get maximum volume

Next things to do:
- Since I've already found two similar nice offers in MundoMedia ("Win a Samsung S6") in Spain I'm going to test them and see what happens. I'll use the best 3 landing pages I've used here (of course after having adapted them to the new offers) plus another one I ripped.
I think that if events will be the same maybe I'll be able to say I'll have found the right placements for sweeps in Spain. This could mean that the next campaigns there may be in green
- I'm considering to launch some low-payouts, directlinking "mobile content" campaigns.

---

Higher bid is not just about volume, it usually also brings better traffic. I say usually, cause it's not always like that.

Yes, when running longer, the differences should be more visible.


Yes, your goal is to get profits, not the biggest volume. But it's still related, without volume, you can't really make good profits. You need to find the sweet spot.

You need a good offer, without it, nothing will help you. Check this thread by flow-rian, took him a while but he was finally able to find the right offer and he is in profit : https://stmforum.com/forum/showthrea...ile-Conquering

Mobile content is certainly a better choice than sweeps - in MY opinion. You will be working with lower volume, since you need carrier traffic, but it's still the best choice for newbies.

---

Thanks for everything Matuloo, I do appreciate you help

I have also read the follow-along you linked and I can learn that testing a lot of offers and spending are the key to skip the line to green

EDIT. Just out of curiosity.
A few hourse ago I launched the campaign I talked about above. 2 offers with a payout of $0,65.
With these initial results

I can assume there is no initial traction at all, isn't it the case to cut it?

One last thing

Would you suggest display traffic instead of pop traffic to promote mobile content? Also if it's directlink?

---

Small test

With the "Win an iPhone 7" campaign in ES of last week I learnt a couple of useful things:
- spend more money and make the campaign run longer obviously give you more data
- thank to this I detected some good placements in ES.

Because I wanted to do the same in another Geo and because there were no more offers in ES and my chosen vertical that were worth a test I did as follows.

I asked my AM for some top converting offers in my vertical and I was given 3 offers ($0,64 payout). I took a look at the offer pages and realized there were already some pre-landers so I asked my AM if those converting offers were running direct-linking or not and I was told they were (direct-linking).

In the beginning I thought that it was a non-sense to test such offers but since they were already converting for others I decided to try anyway.

So I started to test and let the offers run for two days (with the usual evening time targeting of above)

These are the results



Now, also with reference to the post and screenshot above since this types of results are not so rare I'd like to know:

- With this results and after that spend (compared to the payout) do you think it was a good idea to cut the campaign?
- Why do you think these catastrophic perfomances happen even if the offers are converting/have converted for other? Wrong TS? Wrong traffic type?

Anyway, the next time I'll search those offers in Adplexity in order to find the correct TS.

Thanks in advance anybody for help!

ps.
The post previous to this one was a test that was meant to confirm that the placements detected with the last campaign were the winning ones in that Geo (ES) and vertical but unfortunately the campaign was one of worst ever in affiliate marketing

---

Can't see the stats, the attachment does not work.

Offers can work well on other traffic sources or even traffic types.

What works on fb does not need to work on pops.

Even when you're testing "proven" offers, you still need to find one that works well with your funnel and TS.

You can spread your initial test and budget to two traffic sources, to rule out that its just the traffic source that is not performing (each traffic source has some strong and weak GEOs).

---

Fixed .

---

TIP: AM's often don't know much about the traffic they are receiving, some are better at reverse engineering, some suck at it ... some don't case. So take such intel as "affiliates are running directlinked" with a grain of salt. It might sounds weird, but I did better with LPs even thou the offer itself had some LP, same style in many cases. Some users simply need more time to get hooked and that's what the LPs do.

Those results look pretty poor, there is no point in going on with it with the present setup, you need to change something. The question is, what needs to be changed.

First thing I would do is to try some LPs, I rarely run anything without LPs anyways.

Then I would think about the offer, is it a good match for the traffic source you are using? It must be something with a really broad appeal.

Checking adplexity to see where others are running that offer is a good idea too.

Good offer and good creatives, that's the key.

---

I was running them on Popads so I suppose it's a good TS but I often forget to ask to AMs traffic source informations. My bad .

Next:
I've ripped and fixed 3 landing pages (basing on the method Sebastian_r showed in his video ) from adplexity for the next offers. They're waiting for approval by networks now.

I'll update the follow along as soon as it has started running

Thank you anybody for help

---

Home > Mobile > Follow-along Campaigns