Home > Mobile > Follow-along Campaigns

---

From Simple Stormtrooper to General Grievous: Becoming a Dark Lord of Mobile Pops (16)

---

Good work so far! And testing tier 3 geos sounds like a great plan!

How many landers were you testing in the previous camps? Did you drill down to Offer > Lander > OS/Carrier/Browser/etc. to look for any pockets of profits?


This is what I would suggest to do:

If you don't have a winning lander yet, then I would suggest asking your AMs for offers that they KNOW are converting for other affiilates, and test those first. Try to pick a couple that accept the same major carrier, and target just that carrier and test a bunch of landers from adplexity, and cut down to a winner based on the split-test calculator. Once you have a winning lander, mass-test offers either by setting up one camp per carrier, or one camp for all carriers and just specify rules in Voluum to redirect traffic from each carrier to the correct offers.


It doesn't hurt to test 2 sources to see which one will give you more traffic volume or better-quality traffic for your target geo. But if your budget is limited, I would suggest to just run a bit of traffic on each source to get an idea, pick the one you like better, and just do your lander and offer testing on that source. There's little point in running both camps to cut the same offers and landers, spending twice the amount of money to identify a winning lander/offer.


I completely agree. And delayed conversions are indeed frustrating. If that is becoming a concern, try to delay the cutting of landers and offers even when they've reached statistical significance. Run more traffic to collect more conversions, then pause the camp for half a day or however long your delayed conversions usually take to come in. And then evaluate stats and do your cutting and resume the camp. There's never a 100% accurate way of optimizing anything - all we can do is our reasonable best and leave it at that.

Looking forward to your next update!



Amy

---

Thanks so much for your reply Amy!

In my IT campaign, I saw some promise with a particular offer, and specifically targeted certain carriers/browsers that had been profitable in my initial testing phase. I ran $15 worth of ad spend (around 30x payout) on this and saw no more conversions!

That being said, I didn't write down the reasons for my decisions and when trying to retrace my steps now, I can see that it is a bit of a mess. I'm hoping that when I reach this point in another campaign, the Follow Along will help me improve here.

This makes perfect sense - I will definitely try this in my next campaign!

I'm running some new camps today, so I will be checking back in with an update very soon.

If anyone has further feedback on my previous updates then that would be great!

---

That is weird, but is something that happens more often than we'd like for it to. You may have gotten outbid so had the best-quality traffic taken away from you, or there may have been shaving/scrubbing going on, or the offer could have ended without your knowing. And there are many other possibilities as well.

Usually in cases like these, I would compare placements stats before and after the performance drop. If I'm no longer getting traffic from placements that were giving me the most conversions before, then I know I've been outbid. I would bid higher to see if I could get that traffic back and still be profitable. If not I'd just put the camp on pause for a couple days and try again - sometimes the competition would just leave or decrease their bid.


It's VERY IMPORTANT to keep a campaign journal - especially since we're running pop and juggling tens or even hundreds of campaigns! Please see this post for details:

http://stmforum.com/forum/showthread...l=1#post291613

You need to know what you did to every campaign and when. Otherwise you won't be able to assess the impact of every tweak (heck - you may not even remember you DID a tweak!) And you may test the same thing repeatedly, not remembering you've tested it before. (Yup - I've been there, done that, lots of times!)



Amy

---

Good reminder from matuloo!

Your stats look like they're bad - but the fact that lander 3 made both conversions may indicate some hope!

I've given my best suggestion in a previous post:

You need to invest money into finding a good lander first, and then use that to test as many offers as you can find, across as many aff networks as you can afford to run at simultaneously without running into cashflow issues.

Now you have a choice: You could either keep running that camp to cut landers until you have a winner, then mass-test offers. Or just pick another geo+vertical to start testing.

Also - next time you drill down into stats, you need to drill down to several levels.

i.e. Offers > Landers > [something else]

Instead of just looking at overall offer stats, and then overall lander stats, and then overall OS stats etc.

Because all you need is ONE profitable offer+lander+[major traffic segment that has enough traffic] combo to have a nice profitable campaign.

Please let us know how you'd like to proceed.



Amy

---

Here is a thread you can start with regarding testing for bots : http://stmforum.com/forum/showthread...n-Any-Campaign

---

That'a probably just because most of the traffic happened to have come from Samsung.


Good to hear - let's see how that goes!

The most important part of a campaign is the offer, but finding a decent lander first will save you money in the subsequent offer-testing, and give good offers a chance to shine.


You could have for sure, but I'd try to make use of ripped landers as much as possible instead of spending money on translations before I even have a promising camp. I don't know what type of landers you're looking for, but for example if you're looking for iphone 7 landers, you can rip iphone 6/6s landers or other non-iphone7 sweeps landers and edit them for use with iphone 7 offers.


Yup - these are the kinds of mistakes everybody has had to make at least once. Think of it as paying tuition to learn lessons. On the bright side - once you fix these mistakes your camp will look even more promising.


Did you check the win-vector calculator when deciding which offers to cut? I did verify your stats and sure enough, the worst offer is ready for cutting.

The fact that one of the LPs wasn't working would not affect that decision. Landers are expected to perform differently, and that lander just happened to not convert at all - and the reason why does not matter.

Why did you have to separate into 4 camps? Are you targeting one carrier per camp? If so, this wouldn't be a bad approach since you're only running one lander so it wouldn't be like you'd be cutting the same landers in multiple camps. Also, by targeting a single carrier per camp, you'd be able to bid appropriately for each carrier.

The alternative: If you like, you can target all carriers in one camp, and use Voluum rules to direct traffic from each carrier to the offers that accept that carrier.


Sounds great! As mentioned above, try to repurpose ripped landers before spending money on translations.


I don't understand how you'd be losing 50% of the traffic to the wrong carrier - please elaborate and let's figure out how to avoid this if possible.


Looking good!




Amy

---

Are you using any new landers in this camp? Check for logos - propeller doesn't allow logos.

Coincidentally I got a JO iphone 7 camp banned on propeller a couple of months back as well, while testing for the 6-week AMC. And the reason was because of a logo used on one of the landers.

If that's the case with you also, try removing the lander or the logo and resubmitting. If you get banned again, we'll take a detailed look at your landers.


I hear you for sure. There's no need to try to be perfect with anything you do. When in doubt, just do what makes the most sense to you at the time. You will naturally launch and optimize camps better and more efficiently over time. Also, different people have different styles of doing things, so don't feel that you have to imitate everything they do down to a T. If you know of several different ways to do something, either just do what makes the most sense to you at the time, or set up a small test and see which way works the best.

Stats are looking promising!




Amy

---

I realised that there is a small logo on one of the LPs - I've removed that and resubmitted so hopefully that sorts it!

Campaign 2
I had some weird stats yesterday:

yesterday: 834 clicks, 0 CVR
today: 86 clicks, 3 CVR (in line with previous CVR %)

All 3 CVR's came in very shortly after midnight, just before my ad balance on PropellerAds became depleted. It's as if something 'switched back on' at midnight! I've checked with my AM who didn't find anything and said it may be a problem with the traffic source..

For all upcoming campaigns, I'll be doing the following:

• Starting with carrier targeting only
• Day parting 6pm-12am each day

My thinking is that this will ensure I'm testing using the best quality traffic, at the best possible time. I can/will revisit these when I have a profitable campaign that needs more volume.

I'm going to revisit my recent campaigns (AR, EC, PY) implementing carrier-only targeting and day-parting. I had previously written off these campaigns, but I would like to see the difference in results, and as a bonus I may get new hope on these.

---

Carrier traffic is great for sure, but it can also be considerably more expensive. Testing carrier traffic first would make sense though if your offers have a simple conversion flow for carrier traffic.

Looking forward to seeing what results you'll get.



Amy

---

Thanks for the insight. Is the best way to find out about conversion flow on carrier just asking my AM? I am also worrying about volume for this method - for example, the GEOs I am working with only have 5-15k impressions per day 3G. This may not be enough to test quickly (and scale to big profits!)

I’m still having big problems with PropellerAds. I removed the LP with the logo and got another rejection. I had some real promising results with a particular offer, but my progress has been completely halted by this. Every time I go to test, my campaign is stopped.

I’ll be using PopAds wherever possible moving forward as I feel like I've missed a big opportunity here..

The last week, since I’ve had a sniff at a potentially profitable campaign, I stopped launching new campaign to concentrate on what I had. I can’t let this happen, as it kills momentum!

Back to launching camps tomorrow..

---

Yes! But usually the offer description will mention this also. For example this is an offer description I just grabbed from an offer on Clickdealer:

Description
Wifi & 3G
3G - 1 Click
Wifi - pin submit
Telcel
Mexico

So for this offer, visitors on wifi will be presented with the pin submit flow, and visitors using their data plan with Telcel Mexico will be presented with a 1-click flow.

For offers like these, you can for sure justify testing the more expensive carrier traffic from the start. If you want to save money, you could actually start with carrier traffic to see if the offer would even convert. If it doesn't even convert on carrier traffic, or converts very badly, then I wouldn't even waste my time sending wifi traffic to that offer.

For offers that have the same flow for wifi and carrier - they would sometimes still convert better on carrier do to multiple reasons, but if you want to save money, I would suggest running wifi traffic to them first as a test. If they don't convert on wifi, then there would be little point in sending expensive carrier traffic to them.


Is that the total for the entire traffic network? Or the amount your camp will get?

If it's the latter - depending on how well your offer converts, the money may or may not be worth your trouble. I wouldn't bother with anything less than 2k/day even for the newest of newbies, but 5k would be worth a shot.

I have an actual past example of a camp that was getting 15k impressions/day that was doing $400+ in profits/day, and it was a low-payout offer. And yes that's not an example of what would typically happen, but it certainly illustrates the impact of high CR on profits potential.


Could you try submitting a support ticket again to ask for the specific reason?

If they still wouldn't tell you (or give you a canned response), then as a last resort, just run your current-best (i.e. in-the-lead) lander with the offer and keep your fingers crossed.

If you still get rejected let me know and we'll figure something out.



Yup hear you for sure. That's actually a good approach (i.e. to focus on optimizing and scaling stuff that has the most potential at the moment). But since you're encountering rejection problems, it would make sense to start working on something else in the meantime - ANYTHING would be a good distraction as you won't want to keep your focus on the "problem" campaign, waiting impatiently twiddling your thumbs and spewing bad energy all over it.

You're right in saying that propeller is a big opportunity. They have good traffic and good volume. Let's see how you get on with that camp and like I said, we'll figure something out. And in the meantime let's see what new camps you can cook up!





Amy

---

generalgrevious , thank you for sharing your story!

---

Nice work with the AR camp! However, you were drilling down into ISP/Carrier. Could you please drill down to Mobile Carrier instead and confirm that the conversions were from Claro and not "Other"?

Question: Which traffic source is this being run at? If it's PopAds and the tracker cost is posted back by PopAds using the [BID] token, please check the costs on PopAds directly to see what the percentage difference is. The actual cost can be as much as two times as expensive as what's shown in Voluum!

Have you drilled down to Carrier=Claro > Offer > Lander? Are you seeing any green combos? Have you drilled down to Offer > Lander > OS?

Keep running until you're down to your best offer and lander. If the winning offer+lander+targeting(claro) is profitable then, scale to other pop sources. On the original traffic source, clone the camp and tweak the bid to maximize profits, and also gradually cut placements that aren't profitable (I like to cut placements that are in loss by >2x payout).

As for the original camp - you can continue to use it to test landers. Then if you find a better-converting one, replace the landers in the new camps with this one. After finding the best lander, mass-test offers (again, if you find a better offer, replace in new camps).


Is there any particular reason why you're targeting by language?

Especially since you're already restricting traffic by targeting just one carrier, you won't want to further restrict traffic unless you have a good reason to.

I would suggest to target all languages. If, when you drill into stats, you're finding a lot of traffic for a different language than the one used on your landers, you can always get your landers (or even just your best lander) translated, and set up tracker rules to redirect visitors to the proper lander(s) based on the language detected.



This is a common problem faced by many new affiliates.

You'll just have to be polite but persistent in following up. Don't be shy and think "I don't want to bug them too much". The squeaky wheel gets the grease - you need to gently remind once a day until you get what you want.

A potentially-easier way would be to just start every campaign with a bunch of offers instead of 2-3 recommended offers. You can run some initial traffic to the bunch of offers, pick one that seems to be converting the best (no statistical significance needed), and use that offer to test all your landers to whittle down to the best, then use this best lander to retest all offers plus any new ones to identify the best. But even if you were to go this route, you'd STILL need your AM to tell you what each offer's requirements are (e.g. if you can run aggressive), and open up the offers for you.

And, every time your AM tried to push at you what you don't need, just say very politely "thanks I've made a note to explore that later, but let me test this geo+vertical first". It's their job to push certain offers, but yours to decide what to focus on.


There are also other offer aggregators - check out MakeMassive.

You can also start a thread here in the forum to ask for offer recommendations from the many aff network reps that are active here.

Just watch that you're not spreading your revenue over too many aff networks, to the point where you run into cashflow issues (i.e. run out of cash due to balances on various network not yet meeting payment threshold and/or networks not paying out frequently enough).




Amy

---

Just to clarify: You're targeting Arabic in AR? (Insert confused smiley here - the one with the cocked head and question mark.)


Amy

---

No, this applies to my other current campaign (JO). I may have been jumping topics too quickly

---

Ah OK gotcha! Many Arabic geos are still underserved - nice decision to target some of them!

Here are more:

https://en.wikipedia.org/wiki/List_o...icial_language


Amy

---

Home > Mobile > Follow-along Campaigns