Home > Mobile > Follow-along Campaigns

---

6 packs and 6 figures in 6 months (37)

---

Looks like a solid starting position, fingers crossed to achieve your desired goals!

---

Seems you have the right plan Good luck!

---

First of all thank you for the shout-out!

Secondly - nice approach!

A few suggestions:

1)I'm not sure using the same landers you tested for MY for LATAM geos would be the best idea. However, some landers do perform well consistently for multiple geos (e.g. the wheel landers for sweeps). You can use those landers to cut down to the best offers, but it may be worth it to test more landers for those best offers.

2)Another thing to note is that the Spanish spoken in LATAM vs. Spain vs. Mexico are different. It may not be worth it to test this right now, but once you've identified good offers and a best lander, it may be worth it to get the best lander translated into the appropriate languages to see if you could increase the ROI further.

3)I would recommend doing some initial testing on propeller as well - there a lot of traffic there and has become my go-to traffic source for testing.



Looking forward to seeing some stats!

EDIT: Looks like you made a post while I was writing up the previous post! Like I've said - I would recommend duplicating the camp on propeller as well. If you see better results there you can just do your testing there.



Amy

---

Good plan.
All the best!

---

Good point. This is something we're working on creating a unique angle and lander as we speak. We'll test this new lander in our other GEOs.

Okay, we'll give this a shot! Do you think that propellerads has higher traffic quality than Zeropark?

---

Not trying to imply that at all. It's just that different traffic sources are strong in different geos, so doing a small initial test on an alternate source in the beginning can't hurt.



Amy

---

You Sir definitely winning the coolest follow along title thread.

Wish your results to amount the coolness of your title.

Following.

---

Hope that your coming results will be promising...

---

Whoops Argentina was a ice cool shower. Kill that offer.

Yeah, Zero is sometimes difficult, make sure you test both their pop and domain traffic, they can have different results.

As @vortex said, each source behaves differently in different geos, so testing 2-3 for start, to make sure it's really the offer and not the traffic source is not a bad idea (if you have the budget).

---

Makes sense, really great tip. I'll take your advice and split test the traffic source to double check offer potential.

With our current budget, we are going to only going to split test traffic sources on the offers that are converting.

---

Cannot wait to see your results!

---

To further clarify: It's often helpful to set up an initial test on an extra traffic source to see which one has cheaper traffic, better conversion etc. But it wouldn't be necessary to continue testing on both sources while still testing offers and landers - there'd be little point in losing twice the amount of money testing on both sources.



Amy

---

The attachements don't work would you please update them...

---

Let me know if it's working for you guys now

---

Now it is okey..

---

Interesting Journey! Good luck! i will be watching this one for sure..

---

UPDATE

This will be a quick update... we tested these offers on Propeller ads with no success.

We are now moving on to another mass test

---

what do you mean by no success ??? They are worst than zeropark results??

---

Yes exactly, we ran the Offer 1 and Offer 2 from our GEO 1 - All Inclusive campaign on propellerads, and got 0 conversions.

---

Ouch! Stats do look bad. I hardly ever see such bad ROI when mass-testing offers, especially when done across so many geos.

The first question I want to ask is: Why did you use only 3 landers? Before mass-testing offers, it's crucial to have a lander that converts well.

I would suggest to put all your current camps on pause for now. Take your best offer (no stat sig needed - just take the offer that seems to be doing the best) and test 10+ landers - the more landers the better.

Then, we'll see if you can find a lander that's significantly better than the 3 you've tested so far, then decide from there.

Also - at this point, I would suggest to do one geo at a time, just to get the process down first. Attempting this many geos in the beginning may be too big of a challenge - basically every "mistake" you make will cost you multiple times the money you would have had to lose if you were running a single geo.

While testing and cutting landers, there's no need to set up 2 camps to target carrier and wifi separately. It's only when you're mass-testing offers, and only if some of those offers have a simpler conversion flow when run on carrier traffic (e.g. 1/2-click offers), that it would be worthy of the extra trouble of setting up a separate camp and bidding higher for carrier traffic. Otherwise, just bidding appropriately for wifi will suffice, and what little bit of carrier traffic you get will just be a nice bonus.

Nice!! Looks like we think alike! I read this second post after replying to your first post in my previous paragraphs above.

So you're definitely on the right track, and you've seen positive results from testing JUST 3 more landers.

Now go rip a bunch more landers and test them. Cut down to the best, and THEN mass-test offers again.

This time, please do one geo at a time. Let's go through the entire process for a single geo from start to finish. And THEN by all means go crazy with the testing.

You've sustained some losses - but things are looking up!


Amy

---

Ahhh okay... so to summarize your advice I should be testing more landers with my offers. Even after narrowing down the best lander it may be worth it to go back and mass test my offers again but this time using my best lander.

for example.. I do a mass test with 10 offers against 5 landers, from that only 1 offer shows promise.

Then test that 1 offer against the 5 landers + more until I find the very best lander. Then with that very best lander still test the original offers I tested in my mass test as it may show potential with the new lander. Do I have that correct?




As for sticking with 1 GEO.... We'll take this advice going forward. We've just ran a test and it's showing some promise. I'll post the stats tomorrow and we will stick with this GEO until it is fully tested

---

I'm only suggesting that for this particular campaign, because not enough landers were tested.

If enough landers are tested in the first round (using a couple of offers would be enough to cut down to the best), then mass-testing offers in the second round would be enough. No need to go back and retest landers after that. (Although you can of course test new landers, or variations of your best lander to see if you can increase CR further.)


Sounds great and looking forward!



Amy

---

Fantastic!

One thing to be careful about: Avoid comparing lander stats from 2 different time periods.

As soon as you add new landers to rotation, you've started a new split-test. So, when comparing stats across all landers, avoid including stats that were collected BEFORE you added the new landers.

I love to cut corners as much as possible too - but please trust me when I say that comparing offer/lander performance across different time periods can majorly skew your conclusions when trying to find the best offer/lander.



Amy

---

Which is why it can be difficult to recommend a one-size-fits-all campaign testing and optimization approach.

When you're running a campaign, the various decisions you need to make can drive you insane if you're a perfectionist.

Basically just do what makes sense to you at the time, and move onto the next step. As you gain knowledge and experience, you can refine your process to become more accurate and/or efficient.

Again, like you said, opportunity cost.

Also agree that tracking is essential!

Eager to see some stats when you have them!



Amy

---

Update

A lander finally emerged as the winner.

Here are the stats from today's test:




Here are the full stats for the 4 landers since I first starting running them:



Lander 1 is statistically the best.


Next move, I'm thinking about now adding in more offers as wells as making variations of this winning lander

---

Looking good!

I would suggest testing more offers first, and THEN testing lander variations - because the former will typically have a better chance of increasing ROI by a greater percentage. (Sorry for the double-comparatives!)



Amy

---

Will do!

Quick question just popped into my head.

When testing new offers, how confident can I be that an offer is no good if it doesn't convert against this lander? And not that it's just because this particular lander doesn't work with the offer?

I only ask this because some offer pages can look very different


I would completely understand if the answer is that offers just play that much bigger of a role in campaign success than landers and a good offer should convert even with a "bad" lander

---

This is why 1)it would be good to test more landers so that the winning lander is guaranteed to be good, and 2)when testing more offers, only pick ones that are of the same offer type, that is highly relevant for the lander.

It's a very valid concern you've brought up here. In spite of the 2 points above - like you said, even offers of the same offer type can look very different.

This comes down to the same accuracy vs. efficiency argument again. Ideally, if you want to be accurate, you should cut landers for EACH individual offer, instead of cutting based on overall lander stats totaled across offers.

But doing it this way will also cost you. Depending on your situation, this extra cost may or may not be justified.

A good middle ground would be to further segregate offers, and test them in separate camps using the same or different landers, and cut landers in each camp separately.

Take iphone 7 sweeps offers for example. I noticed that there were "test and keep iphone 7" offers and "win an iphone 7" offers. When I looked at landers on adplexity I also noticed there were landers that had the same 2 angles.

In this case, I didn't need to be a genius to realize "hey - let's match up 'test and keep' landers with 'test and keep' offers, and 'win iphone' landers with 'win iphone' offers!" And the results turned out to be quite good.

Another example: Some sweeps offers are equipped with their own "pre-landers", which are essentially built-in landing pages. Say the pre-lander is already a 4-part survey - in that case, putting another survey-style lander in front of it will piss off the visitor. But that very same survey lander may be a winner for other offers - ones that don't have built-in survey landers.

In that case, it would be better to test offers with pre-landers separately, and split-test direct-linking vs. using very simple one-page landers. Even just a js pop-up that when clicked on, will redirect to the offer, coulld work.

Again - lots of ways to approach things. Just some food for thought.



Amy

---

Make sense! Definitely need to pay attention to the nuances of an offer.


As for my campaign. I was only able to find 1 more similar offer to throw in against my winning lander. 0 conversions for day which either tells me; Fridays aren't good days to run in Argentina (data backed this up as historically i've never had a Friday conversion) or maybe the ad viewer is blind my landing page.

Based off this, I'm thinking about doing a few things..

test on Saturday to see if the 0 conversion trend continues.
Test some new landers
Try a new traffic source
and since I'm not able to find anymore similar offers, I may just start creating another campaign to test new offers

---

UPDATE

I've tried running this offer with my best lander for another day and still getting 0 conversions. I'll be moving on to another test.

Here's the strategy. It's a mix of Mr. Payne's and what I've learned from Vortex.

I'm picking 1 vertical to run in 3 - 5 GEOs

All pick 3-5 offers with 5-7 landers.

From there, ill continue running only offers that get more than 1 conversion with the goal to find the best lander for that GEO/Vertical. (will only cut at stat sig).

Once I get the best lander, I may throw in more offers to test against it.

We'll see what happens

---

Sounds good! Keep us posted!

One thing to keep in mind though: When you're testing so many things at the same time, you need to be frugal.

Sure enough, one approach would be to invest money upfront to "break into" all these geos to test each of them very thoroughly until you find the best offer and placements, after which you'll have an edge every time you launch new offers in that geo and TS in the future. But that would also cost you a lot.

If you're going for quick profits to build up cashflow, a better approach would be to test a lot of stuff but NOT thoroughly - basically just spend a bit of money testing each, and ONLY CHERRY-PICK the best that shows promise right off the bat. This is a paradigm shift from testing things thoroughly, but can save you a ton of money when executed correctly.

As usual though the choice is yours to make.



Amy

---

I'd like to second what Amy just said and had to learn the hard way: distributing your resources into many places at once might sound
like a sure homerun but it will actually stretch your dollar far beyond your personal breaking point.

Usually folks do this ( myself included ) out of the false belief that it will work out ( it will but at what cost ) and that you will
be able to reach your desired goal a lot faster ( which is again misleading ) and here the topic is not about how much money
you spend but on how many things at once since they'll all need managing. Without a proper system in place to automate stuff
things will get ugly very fast on pops.

TL;DR
Don't stretch your dollar more than it is needed. Learn the process on 1 traffic source, 2-3 geos, lots of offers, test again until
you get your drivers license. Then, drive on more than 1 terrain ( ie. traffic sources ) and with several cars so your skills develop
right.

Daniel

---

Thanks for the advice guys, definitely going to take a more frugal approach. It fits more into my overall strategy of getting quick cash flow from pops campaigns to fund other tougher traffic sources or GEOs. I'll make sure to only test to reveal the best of the best.

Going off what danielt said, this is exactly what I'm going for I'll most likely lower the number of GEOs from 5 to around 3. Maybe even retest some GEOs from before now that I have a little more experience.

Before I started my mass test though, I decided to quickly test a few antivirus offers that one of my AMs recommended.

Here are the first days stats

2 offers (same payout)
5 landers

(only 1 mobile carrier performed well)



Ran it for another day and cut the bottom 3 landers

Now im testing for stat sig for my 3 remaining landers


Going to dig through the different segments within this dataset to see if I can find any that are performing well. From the looks of it, hours of the day seems to have a few profitable segments (majority of conversions happened at night/evening)

---

Showing some sweet ROI! Verrry nice!

Since you're already in the green, I wouldn't cut anything major unless it looks really hopeless. Instead, I would suggest to keep running to collect more data before making decisions.

Once you've cut down to your winning offer+lander, try to start by cutting non-converting money-sucking placements first. After that, you may find that some of the traffic segments that were in loss before, may be profitable then.

What did stats look like for wifi and the other carriers?



Amy

---

These offers are carrier only and accept 3 carriers.

The stats above are for the best carrier. Only one other carrier got a conversion, we've set up a separate campaign on the other converting carrier!

---

Update

After a good break from running campaigns, my partner and I have decided to switch to FB. We feel that we've learned the ropes enough on pops to be comfortable with the overall methods used in affiliate marketing. Moving to FB was always our plan, we thought it would be best to start with pops as mentioned by others.

We've been reading and following a lot of John Crestani and Ronnie Sandlin's material, and will be following an approach similar to theirs!

---

That would be a very wise plan! It's not easy deciding to leave the familiar to explore new territory, but few people would argue that FB has a lot more potential.

Kudos, congrats, and best of luck!

And thank you for having taken the time to develop this follow-along to page 3 - this is a feat in itself!



Amy

---

Home > Mobile > Follow-along Campaigns