Home > Mobile > Follow-along Campaigns

---

Just popped my AM cherry with sweeps & pops (pun intended)! (19)

---

Quick update:

My first campaign (see above) is dead. I killed it because I had made several mistakes when setting it up: one of my 3 offers was "win iPhone 6" - although all my landers help you win an iPhone 6s. All offers were for Android, but I targeted all devices. My landers needed some polishing.
I'm now waiting for a lazy developer to optimize my landers, then I will start a new campaign with 3 different offers.

Progress: I now know how to set the cost in Volumm (silly me);
I know why there's a difference in the number of impressions from traffic source and Voluum. Click loss! I had more than 20%, though. I think it's due to 3s lander speed;
Currently reading Cashvetizing;
Contracted a web developer, but I have a feeling I will have to "split test" at least 5 to find a suitable professional. I guess work ethic is a scarce asset when working "from home".
Joining as many aff networks as possible to be able to test all offers in my vertical once i have a winning lander.

I should have some results tomorrow, will post an update asap.

Over and out,

C

---

Nicely done Cosmin! You're right - it's easy to spend day after day and week after week reading forum posts, feeling like one is being productive, when actually, without practical experience to provide the context, it would be hard to assimilate a lot of the information into one's paradigm.

TH is an awesome geo, and sweeps is an amazing vertical. Great choices!

When choosing offers: You'll see some really low-payout offers in the $0.10-0.20 range, and also offers in the $1-2 range. A few months back I did some testing and had absolutely no luck with the $0.10-0.20 offers, but had success with some of the $1-2 payout offers. YMMV, but I would suggest that you include more $1-2 payout offers in your testing.

(Note: There are low payout offers that convert great - please don't let my suggestion above discourage you from testing low-payout offers in the future.)



Simply add the [BID] token in your Voluum's traffic source settings for popads:



Don't forget you'll need to re-copy the campaign link and replace the destination URL on popads with this new link for costs to be tracked (because the new link will contain this new token).

IMPORTANT: The costs sent back to voluum will NOT be accurate in most cases! Always double check popads stats (within popads' user interface) to find out what the actual costs are. The costs posted back to voluum will often be less. Worst case I've seen is when the actual costs were twice as much as what was reported - this means I'd need to see 100% ROI in voluum just for the camp to be breaking even in reality. So be mindful of that trap.



$30 spend with a single conversion doesn't look promising for sure. But how much were you bidding?

You'll want to bid above-average in order to avoid getting crappy traffic, which there's a lot of especially when running pop.

Secondly - 3s lander loading speed is dreadful! Caurmen has told us that anything beyond 2s will cause you to lose visitors in droves.

My guess is that you're not using a CDN?

You'll find all the information you need and more in this section:

http://stmforum.com/forum/forumdispl...verything-CDNs

Also - this post contains a list of threads that can help you decrease the load time further (in the section: "Try decreasing your page load speed!"):

http://stmforum.com/forum/showthread...-AM-FAQ-Thread!

Regarding ripping and fixing up landers: If budget isn't an issue for you, and you can't be bothered to learn coding yourself, I would suggest recruiting a part-time freelancer from freelance job sites such as upwork. If you need a job posting example for this please let me know, and I'll post it here.



I would definitely bid above-average (if you weren't before), and would definitely fix the page load speed issue, and then re-test.



I'm definitely not an expert in money matters, but having a USD account would be wise. You can arrange for all aff networks to wire your commissions to this account, and can also pay traffic sources from the same account, so that everything will be in USD.



Great steps to take!

And yes - knowing coding basics yourself will help immensely even if you're planning to outsource.

You're doing wonderfully well for someone who had close to zero experience with online marketing before this! There's a huge learning curve and you're making amazing progress. Keep on keeping on and you'll get to where you want to go.




Amy

---

Amy gave you good advice, I would follow it.

Whether the low .10-.20 offers will work well in TH right now, I'm not sure, but I have had good repsonse in other geos for those type of offers. Properly hosting your landers will be even more important due to the volume you will need in order to make up for the $1-2 offers.

---

Oh yes absolutely! I love low-payout offers. I once promoted a 0.12 offer that was doing 3 figures/day in profits the first day. Thanks mrpayne for emphasizing this point!

What I said about $1-2 payout offers working better than $0.10-0.20 offers - I was referring specifically to TH sweeps, and only based on my limited experience with this geo+vertical (I think I tested maybe around 20 offers.)

So definitely don't let me turn you off from testing low-payout offers!



Amy

---

Yea, I understood what you were meaning. I just wanted to show that there is definetly potential with both types of offers. Prevoiusly I would never have considered offers less than .15 because I didn't know how to approach them.

Now, I also like them very much. The second campaign I had profitable in my follow-along (before the payout got slashed) was on an .09 offer.

@csstaq - Let us know how your offer testing goes!

---

@Amy - thanks for the input!

The 2 landers were chosen by spying what LPs got most traffic for a certain offer in that geo. But ya, i should test at least 5.
Regarding the bid, I'll use average+50% and when I have winner Lander+offer will use stag bid.


Problems:

- Voluum is showing errors, equal number as conversions. I'm probably doing something wrong...
- Conversion tracking - I didn't manage to send info back to traffic source. The postback url ends in: id=[IMPRESSIONID]&bid=[BID]. Moreover, I didn't manage to get the cost from the traffic source. And I feel like part of my brain is melting. Amy, could you, please, try to explain (again) how to track conversions between Voluum and Popads? Imagine you would have to teach a monkey how to do that. That's where I'm at right now. Thank you!
- daily budget ($45) gets depleted in approx 8 hours, although I've set Automatic throttling - Maximal spending - $3.6 per hour. Is there any way to spread the spending through the entire day and still get reliable data? Or should I just spend more per day?
- click loss; I'm losing 35% of the clicks (i.e. popads - 10000 impressions vs voluum - 6500 impressions ) while LP upload speed (with tracking) is 0.65s. It should be around 20%, right? Is there anything else I can do to gain more clicks.


Progress:

- currently running camp 1LP - 2 offers, got about 6 conversions, -75%ROI. Will post results later tonight.
- CTR is very low, around 12%. The lander I dropped yesterday had a much better CTR, around 80%, but no conversions.
- both offers are converting (offer1- 4 cv & offer2- 2 cv).


Questions:

I would assume (and I don't like to assume) that both offers are worth the struggle. Should I continue by testing then separately, in diferent camps?


Next:

Will look into how to cut placements
Test more offers with this lander, but not so sure about CTR, only 12%...I guess I should go back and test more landers.


Next update coming asap.


Thank you,

C

---

Focusing on lander variations is a good idea - but make sure they're big variations.

A classic mistake in split-testing is to split-test tiny changes - font size from 12pt to 11pt, change in the shade of colour of a button, etc.

Test the BIG stuff.

---

Ah OK I see! If you find that it works well for you, then stick to that. Maybe after a few tries if you see that landers that were getting the most traffic are turning out to be winners consistently, then you can save test budget in the future by only including high-traffic landers.



Huh? Equal number as conversions? What do you mean?

It's probably due to the postback not having been set properly. Could you post here the offer link you entered into Voluum, as well as the postback link you entered into the aff network? I'll take a look for you.


You mean cost not conversions right?

It's weird that it's not working for you! It's really straight forward. Basically, it's just 1)adding the BID token into the traffic source settings in voluum, like in my screenshot above (a couple of posts ago), and 2)replacing the campaign link on popads. My campaign link looks like this:

http://voluumdomain.com/94f35128-0b1...3548e80db2?v1=[WEBSITEID]&v2=[KEYWORD]&v3=[COUNTRY]&v4=[CATEGORYID]&v5=[ISPNAME]&v6=[DEVICENAME]&v7=[OSNAME]&v8=[ISPID]&v9=[BID]&v10=[QUALITY]&BID=[BID]&ClickID=[IMPRESSIONID]

You don't get conversion data from popads - it wouldn't know which offers have converted. Only your aff network can send that info back to Voluum. Perhaps that's where the confusion lies?



I wouldn't worry about spreading traffic throughout the day to get reliable data.

Since your current goal is to split-test stuff (offers/landers), the day to day and hour to hour fluctuations will impact all split-test candidates - at least that's the assumption I'm making. You could either spend more per day, or just implement dayparting, and restrict to hours that typically convert the best, i.e. after work hours (e.g. 5pm-12am) or also include work hours (e.g. 9am-12am). Or if you really want to spend less than that, run during a subset of the evening hours - e.g. 5pm-10pm.

You can always open all the hours back up once you find a promising offer.


I'm afraid I don't have a solution to this - if you ever find a way to decrease clickloss please let me know as well.



I can't think of any justification for doing that. What's yours?

This would not be the right time to focus on cutting placements - if you see really bad ones that are draining your budget and not converting, by all means cut them. Your current focus should be on testing offers and landers!

Also - would be nice to see some data for Offer -> Lander -> OS - I want to see if there are green combinations.




Amy

---

Heed caurmen's advice and you can't go wrong.

I don't even think you're at a point where you should be testing lander variations. You've only tested 2 landers so far. I would rip and test more landers, either use your current offers or a few new offers to test them to identify a best lander, THEN use that winning lander to mass-test every similar offer I could find on all aff networks I'm with - that's when the magic usually happens.

Your current offers aren't converting very well - but I have no way of knowing because it may be your landers (because you've only tested 2). If you like, you could ask your AMs for some new offers to test your new landers with. Or, you could just use your current offers to cut those new landers.

Either way, your first step is to test more landers that look different from your current ones. Once you have a good lander, mass-test offers.

One of my favorite sayings is: A good offer will convert on just an OK-lander, but not even the best lander will convert a crappy offer. Therefore, I usually suggest using ripped landers to mass-test offers, and THEN only if you have some profitable camps, to go back and test lander variations etc. It takes work to make lander variations, or test original lander ideas, or variations of lander angles. Reserve that kind of heavy-duty work for the profitable campaigns.

Try not to invest too much time on initial test campaigns. Cast a wide net, test lots of stuff using minimum time (i.e. using ripped landers for example), and THEN invest more time when you've found a really promising offer. That way you're not tying up all your time trying to optimize a camp you have no idea will ever even make you profits.

Sorry for being long-winded!



Amy

---

Hey -

Just wanted to say as a noob myself, you're doing one hell of a job diving head first (i like it). Inspiring me to create my own follow along and not feel embarrassed about asking stupid questions.

Love the community here as this thread is a prime example of the friendliness from others. I'll hush now and sub -

Cheers mate.

---

^^ Lolz, good to hear that. It's not easy to get over the fear of failure, I have to say.

But, you're never fully prepared to do new things.

We're better off by just pulling the trigger.

Making mistakes is called learning.


Thanks,

C

---

Yes, yes, absolutely, yes.

Newer affiliates reading this thread - heed this advice above all else.

Job #1 is ALWAYS finding the promising offer/traffic source combo.

---

@Amy & Carumen

Just in case you need validation:

I took out the " free gifts" text from the lander and ran a quick test with my 2 "good" offers. Not a single conversion. I hope this reinforces (in my head) the fact that you have to test test test test test until finding that good offer.


Yup, the cost is now showing in Voluum!!! Thank you!

Don't really know why I wanted to post conversions back to traffic source... I guess it makes no sense, now. Lol.

I don't know what the errors are saying, how can I know?

-------


I've posted the job offer on Upwork, will update you in due course. What countries should I target, what do you guys suggest? I've read India is a no go. Same goes for Pakistan?

And right now I'm searching offers, spying and procrastinating big time (booooo!). I need to find a reliable developer asap!!!


Thanks,

C

---

Yup that's inline with my experience as well: Rip enough landers and they'll be good enough to uncover good offers for you. But if you have a dud offer, then not even the best lander in the world will make it convert.


YAY!! I'm so relieved - cause I was all out of ideas lol!

If you're still getting those errors, send a support ticket to Voluum support - it's SO easy to do directly through the interface, and they have great response time.

Oh it does make sense that people will sometimes want to post conversions to the traffic source. This way you'll be able to see both revenue and cost information TOGETHER, for each placement for example. I'm a bit paranoid about this though - many traffic sources have their own internal media buying team running offers and competing against us. We really don't want to make it TOO easy for them to spot profitable camps do we

Instead, I prefer to export placement costs from the traffic source, then export placement revenues from the tracker, and combine them using excel formulas:

http://stmforum.com/forum/showthread...e+cost+revenue



Sounds great! I haven't had much luck with IN or PK either (except when targeting carrier traffic). If you do target these geos, cut placements VERY aggressively in the beginning if you don't want to lose your shirt.

Many geos in Asia, LATAM and Africa are great. TH, ID, MY, SG, DK, AR, BO, LK are just some of the best ones I've run in. You may want to write these down in case I change my mind about being so generous and edit them out later.



Amy

---

Amy, you're amazing, thanks for all the support!!!


I was just wondering, isn't there a tool for optimizing LPs, to save me from learning code/hiring developer?

Then I came across Lander Ds (beta). I'm gonna give it a try. IF anyone is using it, please let us know your thoughts.


Thanks,

C

---

You should edit them out

---

Lander DS looks really interesting! I just watched the youtube vid. Does it only optimize the page for loading speed, or will it fix all the errors as well?

Do let us know how this works out for you!



Haha I was just joking. Every geo has potential to make you money, and success will mostly depend on how many offers you test. Just because I've had success with these geos, doesn't mean other people will. So - get testing!



Amy

---

Haha I was just teasing, as I've had good results from several of those geos and including IN.



Andrew

---

One thing that's worth noting here: for some reason, different people will find different geos and different offers radically easier or harder to convert.

Some offers are just trash, obviously. But even amongst good offers, you'll see time and time again that some people can convert Offer A without even trying, wheras they just can't make Offer B profitable, and other people will be saying "fuck man, Offer A's trash, go for Offer B, it's easy money!"

The same applies to geos. For example, personally I struggle with France as a geo. Don't know why, but despite knowing the country moderately well and speaking the language moderately badly (advantages I don't have with, say, Brazil), I still find that I work far harder for a conversion there than I do in other geos - and in other geos I find I can get conversions comparatively effortlessly whilst other people find them a pain in the ass.

---

Home > Mobile > Follow-along Campaigns