Home > Mobile > Follow-along Campaigns

---

First voluum campaign (18)

---

Hey

I have been doing marketing affliation for a few years, but was only promoting on one market and one product. So we built up everything ourselves for tracking etc...

Now we are going a step further as we would like to spread out, for this, i joined stm a few months ago, learnt about affiliate network, sources, tracking etc, and now we are up with our first "real" campaign !

So here is the stuff :
Affiliate network : f5
Vertical : sweepstakes (win something)
Geo : some eu country
Payout : 0.40$
Trafficsource : propellerads
Tracking : Voluum

So yesterday (09/16) was our first day, and it took up some time to setup hosting, cdn, cname, find an offer, find a good looking landing page, edit it and start the campaign. Our goal was to setup the campaign before midnight and see the first numbers, just to make sure our setup was correct and that it was really working.

So the results are the following :
Visits : 16 436
Clicks : 542
Conversions : 14
Revenue : 5.60$
Spent 54.62$

Some might think it s catastrophic, but we were pretty happy doing conversions as it shows it might be working
This first campaign showed a few problems :

About Voluum :
1- We couldn't find out how to post back conversions from voluum to propellerads and thus couldn't delete bad zones
2- We didn't track the cost of the campaign in voluum
For both these problems, i contacted voluum and hopefully will get an answer from them to solive this issue !

About affiliate network :
1- I tried to get in touch with my am to get him tell me which offers/geos are doing great at the moment. But his answers were a bit bullshit like "ask your traffic sources what geos are good to work with cheap prices"
Well ok, but i don't want to work according to the source, i want to work according to the network... am i wrong there ?
Anyone can recommend a good affiliate network with more helpful people ?

About landing page :
1- We did a pretty basic landing page, so simple quizz leading to the offer. Now i read here and there, that some were sending one surfer to multiple offers, how should it be done exactly ? (back button ? popunder on each quizz click ? ...)

We will try to launch a new campaign this evening with the small datas we gathered from the first campaign, hopefully for 50$ spent we hope to reach 15-20$ revenue, but we will see !

thanks to anyone bringing some help, and feel free to ask additionnal informations if needed !

---

1. You don't need to post back conversion to find out which zones are not performing. Click "report" under the campaign and you should see a tab that says variable. Click on that and one of the variable should be a website id or zone id. Than you can compare your cost for website id in propeller vs the revenue in website id in voluum. Generally, a rule of thumb is not to share your data with the traffic source.

2. Sometimes you cant track cost accurately, specially with pop with so much click loss. Best is to use a "update cost" for the previous day.

---

Ur AM advice isn't BS. I see you are running sweeps in propeller, what made you run sweeps in propeller? did you manually spy someone running sweeps, did use check adplexcity and saw many sweep lander? It always smart to ask the traffic source what kind of offers do other affiliates run, it gives a good sense of the audience/quality. If you want to run sweeps, mundo/CLickdealer/admislls/affliaxe are good options but I am not sure if they take on new people.

Backbutton does wonders if you can figure it out. Usually sending someone to the offer page is not a good idea as a backbutton.

---

thanks for all your tips

For Voluum you are right, i see zone properly and could blacklist some of them on propeller.
We randomly choose the offer, just to start somewhere.

I didn't expect to have such information from the traffic source, so i ll contact them to see what they answer, wouldn't have think about doing it, thanks !

---

Nice start for sure!

Rick you're right - for a first camp, $5 revenue on $50+ spend isn't catastrophic. Heck, just having the infrastructure set up correctly, i.e. hosting and CDN and tracking etc., must have been a real challenge. It sounds like things were set up correctly. So let's discuss testing methodology - some suggestions:

1)You need to be testing a LOT more landers than just one. Do you have a subscription to Adplexity? If not I would highly recommend it - it has become an indispensable tool for many affiliates.

http://stmforum.com/forum/showthread...e-STM-Discount

Rip at least 5 landers for your vertical and geo, 10 would be better. Pick ones that look different from each other and not just different in minor ways.

2)Join more aff networks - the ones suggested by webdev are great. You need to test multiple offers. Would be good to ask AMs to recommend some that are converting for other affiliates. Basically you need at least one offer to convert well enough to help you cut landers down to the best one.

3)Once you have a winning lander, you can test ALL similar offers you can find from all aff networks you're a member of.

4)Also - when it comes to setting bids, try to test either 3 staggered bids at the same time (low/average/high) and run 10-20 times payout to each, to see which ones give the best ROI, and use that camp for the rest of your testing. OR, just run one camp but make sure you're bidding above average - 50% above the average bid for example. You want to make sure the traffic quality is there, so that if performance is bad, you'll at least know it's not because you're bidding too low and getting junk traffic.

This is basically it for a start. If you read a couple of the recent and longer follow-alongs for pop, you'll see that members that have seen the most success, are the ones who are willing to cast a big net over a wide area, i.e. testing many landers and offers (ESPECIALLY offers!)



As for passing costs back to the tracker - most traffic sources don't have a token for doing that, and the few that do, are often not accurate (with the exception being Voluum+Zeropark but that's only because they're from the same company Codewise). It would be better to export placement cost stats from the traffic source, export placement revenue stats from Voluum, and combine them using for example this spreadsheet here:

http://stmforum.com/forum/showthread...kly-amp-Easily

Yup - you can add the backbutton script, which has become a standard basically. Some affiliates will take the visitor directly to the offer when they hit back. Some will redirect visitors to a monetiser service (e.g. afflow/YTZ). Some will make another/several landers available for the visitor to back up into - the most aggressive will arrange them in a loop so it's neverending (warning: this kind of tactic can get you banned from traffic sources so please be careful).

And yes - some will code the lander to spawn a new popunder when a click is triggered (e.g. when the visitor clicks on the CTR).

Here are some posts with the lander tricks you mentioned plus more:

https://ppcmode.com/landing-page-code/

http://stmforum.com/forum/showthread...ts-for-landers



Will look out for further updates from you!




Amy

---

@webdev what would be the reason behind it? is it just for when issues occur or is there any other reasons?

---

thanks vortex !

Yes we do have adplexity and we ripped our first lander from there !

Question about back button, on click etc... do you track those in a different campaign, or is there a solution with Voluum to track everything in the same campaign ? Couldn't sort it out with click/1 click/2 etc... (and it s more about split testing offers than having multiple offers on back button etc , ie : main offer = click/1, backbutton = click/2, onclick = click/3 etc...)

I think our major mistake atm is to be on only one affiliate network. We should at least join 3 or 4 i guess + things look to be slowed down a lot on week-ends (offer approval etc), i hope we will move forward faster on monday
I ll keep the topic updated as soon as we have more stuff done !

---

Today's stats

Visit : 17 221
Clicks : 14 581
Conversions : 20
Revenue : 8$
Spent : 25.91$
ROI : -69,12%


- Changed LP -> increased the CTR nicely
- Reduced bid (more than /2)

I just launched a new campaign with specific browsers/OS etc.. from the stats i could gather above, will post the results tomorrow.
Also, i ll focus next day on finding new offers (seems to me this one is not good, payout too low, not attractive signup page, proof is CTR increased a lot, but CR didn't etc...), and get new landers, so that i can split test 3-4 landers and 3-4 offers

---

Make different campaigns for backbuttons and any other autoredirects, you don't want those to skew up the main campaign data. When doing the optimization, you need to focus on the initial performance mostly, redirects will be kinda like a bonus.

More aff networks are definitely needed too, don't start to use too many straight of the bat thou. You would end up with a few bucks at each of them which doesn't help the cash-flow at all.

---

thank you for the tips

---

Still the same offer, improved a bit the lander, and focused on some brands/os

Today's stats

Visit : 5 565
Clicks : 7 560 (uh, backbutton and popunder were tracked Under same campaign back then))
Conversions : 10
Revenue : 4$
Spent : 8.12$
ROI : -50.76%

I think with some more work and adjustements i could get to -20 or -30% ROI, but the offer really seems too low on payout compared to the other offers on the same geo

I have send my application to affiliaxe, adsimilis and clickdealer. Will see if they accept me and what kind of offers they could show me !

I m stopping that campaign with that offer, and until i get an answer for the above networks, i m going to try some a4g offers. Will post stats !

---

Very nice!

Have you drilled down to OS and carrier?

I would suggest to hold off on scaling for the moment. How many offers and landers are still in rotation at the moment? It would be good to have a winning offer+lander+targeting combo first, and then scale.



Amy

---

For OS and carrier, it s split in between many different, so really impossible to say atm
The only thing i could do is cut tablets and two zones. And i only keep one lander.
I increased the bid a (very) little bit to try to still get some volume !

I ll post results tomorrow !

---

Just want to make a clarification here that when it comes to traffic segments, you do NOT need to identify the best segment - for example the best OS or carrier - and cut the inferior ones. You will want to keep running as many segments as you can as long as they are profitable for your best offer+lander+bid combo.

(Only when it comes to the offer and lander will you need to cut the inferior ones - because you can only show a single offer and lander to any one visitor. Whereas traffic segments such as OS/carrier/placements are NOT mutually exclusive. The more profitable traffic segments you have, the more money you make.)

And when you're still testing offers and landers, and traffic volume already isn't too high, then you really should cut sparingly or not at all unless a traffic segment is really not converting. You need that traffic volume to complete your testing.

However - if you're getting a lot of traffic for your targeting options, then you can afford to cut aggressively any traffic segment that doesn't look promising, to target only the best-converting segments to help you test for cheaper. For example, say you're running in a geo like TR or BR, where you're getting tons of traffic. You're seeing that Android is converting a lot better than IOS. In that case you may just target android when testing your offers and landers, to test for cheaper and get green on Android faster. (Something to note though, is that by targeting only Android, any new offers or landers you rotate into the camp, you'd only be testing for android - i.e. the best offer+lander may not work as well for ios. Therefore you may want to re-test offers and landers for IOS later.)

I'm a bit long-winded here. You may or may not be making this mistake, but I thought I'd clarify anyways because it's a really common trap to fall into for newbies (been there and done that myself).

Looking forward to seeing your results tomorrow!



Amy

---

So, not much too say in fact, i m at -45% ROI
However i fucked up the beginning with : instead of excluding a zone, i set it as the only zone to send traffic, ahah...

anyway, i ll let it run until tomorrow evening, and post stats tomorrow, and see what happens !



I m still testing many countries/sweeps offers in the meantime and results are all pretty poor (even had a 0 conversions over 58$ spent on an 1.5$ payout offer)

I m working with clickdealer only at the moment, and will keep going for a few more days, I hope my am will give me some good infos on monday .. who knows

Also, I m still a bit confused on how to find offers that really generate money. I can forecast that reaching 5-10$/day with an offer/geo is someting I can do, but I don't see how an offer/geo can lead to xxxx$/day... or maybe it s because sweepstakes don't allow that ?

Any comments are welcome

---

You'll get MUCH better results if you:

1)Test lots of landers for that vertical+geo to identify the best lander.

2)Use that best lander to test all the offers you can find across at least a few big affiliate networks.

Step 2 is usually where the magic happens, but sometimes you'll start seeing profits in step 1 - depends on how lucky you are with the offers you used in step1. So even if you see green in the first step, as long as you're getting enough conversions to cut landers, you may want to keep spending money to cut down to the winning lander, just so you'd be able to mass-test offers in step 2.

Casting a big network is almost the only way to profit consistently in pop - unless you have some edge that other people don't.

Clickdealer is a fantastic network, but DO sign up to at least a couple more!

Mr Green is doing mid 5-figures/day in sweeps:

http://stmforum.com/forum/showthread...l=1#post285360

It's about testing LOTS of offers and landers (ESPECIALLY OFFERS!!!)

So - throw your doubts out the window, keep the faith, and keep testing - we'll get there!



Amy

---

Quick update, I m still stuck between -30 et -80% ROI on several campaigns. However, it looks like the campaigns i m running also depends on hours of the day and days of the week

I ll keep testing on this vertical.

I joined the 6wamc, so until then, i ll stick to it, and keep looking on what can be improved little by little !

---

Try not to replace your offer and lander testing efforts with the cutting of placements and hours of day - not saying you are. Just more of a heads up.

Test all the different landers you see on adplexity, and all the offers you can find for your vertical+geo, to get your ROI as high as you can BEFORE cutting placements and doing dayparting. The process is much less expensive this way, and your scaling efforts will result in a higher percentage of traffic sources being profitable.

See you inside the AMC!


Amy

---

Home > Mobile > Follow-along Campaigns