My 40-day tutorial with Voluum - Let's go ! (5)
Hello everyone, I started the tutorial around two weeks ago, and this is where I'm currently at...
Affiliate networks - got approved by
Traffic networks - PropellerAds and PopAds
I launched my first campaign on PropellerAds... no conversions at all.
Then I launched my second campaign on PopAds this weekend, and just an hour ago, I logged into
Now that the first step is done, it's time to pick another offer and work on those landing pages.
Hello!
It's great that you started a follow long. Would be nice to see more information about your campaigns. If you have any questions or want some tips - feel free to ask 
Now that the first step is done, it's time to pick another offer and work on those landing pages.
It's been 2 weeks - have you had fun with landing pages yet?
Amy
Hi everyone, time for an update... I finally started my first multi-offer/multi-lander campaign.
- First I chose to work with the sweeps/iphone vertical, and picked 2 malaysian offers on Mobidea :
- Then I downloaded a dozen landers from Adplexity, went through the Codecademy coding courses as recommended, which took a bit of time and started editing 2-3 landers... Got a bit lazy over the past 2 weeks, until I saw the email from Adplexity about the landers megapack for just $1, so I quickly grabbed it - no need to fix landing pages anymore !
- In the meantime I also set up hosting and CDN - ran into the certificate problems but solved it thanks to all the useful explanations in the related threads.
After that, I created the campaign on
- I made the small mistake of putting the single-offer click url inside my landers instead of the multi-offer one, so the first offer go twice as many clicks as the second... Anyway I had also set up
- After all these visits and no conversion, I paused it after one day, so my question is : how much traffic does it take to determine if the offers are good or not ? Are these traffic numbers enough or do I need a lot more ?
- By the way I have reached my limits on the AWS free plan so I'll have to upgrade - any advice about this ?
- Regarding budget, I went with $10/day again, but how do you determine the budget you need for testing ? I saw some kind of formula with payout, number of offers and number of landers, is that it ?
- I used english landers, but since it's Malaysia, I should probably test at least one lander in Malay language ?
Anyway I'll have to read and re-read and take lots of notes regarding testing and optimisation...
Edit : ok I found the answers to my questions by reading the day 30-32 lessons, it's much clearer now. I also applied to more affiliate networks using the STM application tool, and next week I will set up more campaigns and test other offers, landers, geos...
Yikes don't know how I missed your post!!!
Next time, if you don't get a reply in say 3-5 days, please bump it. Profuse apologies for having missed your post!
However - the reason why I've not been spending as much time on the forums as I normally do, is because I've been working on something that may help you since you've been playing with sweeps offers:
https://stmforum.com/forum/showthrea...268#post378268
The rest of the guide will be posted over the next few days - so stay tuned!
https://stmforum.com/forum/forumdisp...age-Depository
If all you're wanting to do is split-test offers, that would not be the way to do it.
You'd only need to add a lander to the tracker as a multi-offer lander, if you're wanting to show more than one outgoing links on your lander that would EACH go to a different offer.
Example: Your lander shows 2 sweepstakes prizes for the visitor to choose from - iphone X and samsung galaxy s10. When they click on the first link they're directed to the iphone X offer, and if they click on the second link they're directed to the samsung galaxy s10 offer.
But if all you're wanting to do is split-test the 2 offers, i.e. visitor 1 to your lander will click through to the iphone X offer, visitor 2 to your lander will click through to the samsung galaxy s10 offer etc. - then you would:
-Add lander to tracker as single-offer lander
-Add each offer to tracker
-Create campaign in tracker, and choose the lander, as well as both offers
As for
How much are we talking about? Do you have a link to the pricing info?
-You don't want to spend too much money on testing an offer if there are a ton of offers available in the niche you're testing. Sweeps offers for example are a dime a dozen, so you'd probably want to spend less money testing each, so you could test more offers for the same amount of money. Don't spend money until you're 100% sure the offer is bad. Great offers usually perform well from the start. Exception: An offer that came highly recommended by someone you trust (another affiliate, an AM you've been working with that has been sending you good tips) - for those, by all means use every trick in the book to give it your best shot.
-On the other hand, you don't want to spend so little that you don't give the campaign a chance to start performing. For example if you're throwing 5 offers and 5 landers into one campaign, and the average payout is $1, you can't just spend $10 and expect to get even a rough idea on performance.
-So then what would be a sweet spot? That's where experience will come in. But even the pros will sometimes overspend/underspend. Most of us aren't psychic.
-It's always harder to evaluate the performance of a single thing in isolation - which is why whenever possible, don't just run one lander with one offer. When you're split-testing multiple candidates, you can use statistics to weed out the worst candidate. And if you're lucky enough to have at least one good offer and one good lander in the mix, then you're off to the races. This way you know for sure exactly when to cut something - just follow the stats calculator. (Instead of looking at the 1 offer + 1 lander and thinking "should I stop running this?").
-However: Beware of the situation where all the split-test candidates are losers! That can happen as well. This is where that formula you were talking about can save you. You set a maximum amount of test budget you'd be willing to spend initially, for example # of landers x # of offers x 2 x average payout. So using the same example above of 5 landers, 5 offers, and average payout of $1, you should set a budget of $50.
After spending the $50, if you're not seeing a healthy number of conversions from any offer, it may be good to just swap out the entire batch of offers and replace with new ones.
And if you ARE seeing some promise, then add more budget and start cutting landers/offers/etc.
It doesn't mean you always want to run to this maximum budget before deciding whether or not to keep running. For example, if I know my landers are solid, and I only see 0-3 conversions made by different offers after the camp has spent $25, I'd probably stop the camp.
-But then it would depend on what kind of geo I'm testing as well, and whether I'm targeting good placements.
Example: If I'm running a US camp and know I can get a shit ton of traffic volume, such that I'd be able to cut heavily and STILL make bank, then I would just suck it up in the beginning and keep throwing money into the campaign until I have a "winning" offer and lander, then throw in more offers or landers for a 2nd split-test, then a 3rd - until the continuous-improvement cycle gives me what I feel is a great offer+lander combination. Then I'd cut heavily to make the camp green.
Another example: If I'm running in a geo I already have a good placements blacklist for, then I would set a much lower initial budget for testing, because I'd expect conversion rates to be high right from the start.
-Always make sure your landers are solid, and that your traffic isn't crap (i.e. use a trusted source where you're confident about the traffic quality, for your testing).
There are so many considerations involved in a decision like 'when to stop a campaign', it would be impossible to cover them all, plus every situation is different. But hopefully I've provided some examples to give you an initial understanding to build further on.
I personally wouldn't bother testing other languages until I've found an offer I deem at least somewhat promising.
You can target the English language (browser language) when setting up your campaign on the traffic source, and use English landers. If you see good results, can always split-test other languages.
So what are you running currently?
Amy
Home > > Newbie Follow-Alongs