Budget to be set aside for TESTING based on each offer payout (8)
Budget to be set aside for TESTING based on each offer payout
Hi Guys,
I believe most newbies fail because they do not dare to spend more money to test!
For example, for offer payout @ $2.20 and when they test with overall $30 with no conversion, they concluded that the offer cant convert.
However, the case is it must depend on whether each placement/targets have spend at least 2-3 times the payout offers to justify if the offer cant convert.
I have a question for all the experience super affiliates.
Let's take for instance Offer A and Offer B.
Offer A payout is $0.30 while Offer B is $2.50.
How much overall budget would you set aside for testing Offer A and for Offer B? (How do you come out with that budget?)
Thanks!
I don't have a set budget "formula" , I do it based on feeling honestly... lots of factors play into it. Absolute minimum is around 10 times payout for me ... usually more, depends on the offer and payout as well.
For something that's in a vertical I know ... where I got proven landers from previous campaigns - $50 testing budget for a $2.5 payout offer is more than enough to determine if it's worth investing more. The data will show if just one shit placement ate everything up, or if the GEO or traffic source overall is not worth it ...
If it shows potential maybe around -50% ROI or better off the bat , I usually invest quite a lot of money and try to make it work for quite some time. It happened to me several times already that I let go after 2 days giving up on the offer, only to return to the same offer two months later and make XXXX a day off it ...
When I was starting out the biggest issue was a lack of reference experiences. I had no idea if I had spent enough , if just the GEO was shit , or the traffic source , or the offer. Now I know the traffic sources and GEOs , and if I test a new offer I know it must be the offer .... in the beginning it is pretty tough , but you'll get those reference experiences soon. The learning curve is pretty steep in the beginning ..
Originally Posted by kepe95
The easy part would be to determine if the offer has any potential. We can simply ripped off landers from spy tool and spend money for the data.
I guess the hard part would be to turn offer that shows potential into xxxx a day.
I know alot of general guides have mentioned how to optimize our campaign to increase the revenue. Do you mind to share your general work flow (without revealing any of your main secret) on how you usually do to improve any potential offers from negative ROI to xxx or xxxx a day?
I will run very broad at first, even if I do not expect anything. Often something like tablets and iOs falls away quickly, and the ROI gets much better instantly.
2. Only then I start cutting the unprofitable stuff. If I want to reach higher volume I'll try to keep the targeting as wide as possible.
3. Never stop split-testing. Getting payout bumps ...
Usually it is mostly about finding the right offer and GEO combination nobody is aware off yet , and matching it with a strong lander / angle .
I made some mistakes in the past, where I cut half the carriers during initial optimization, and once I was profitable I had lost half the volume from all the carriers I had cut. They were not profitable in the beginning, but would have been so after everything else was properly split-tested and optimized ...
While I don't recommend it I start all new campaigns with no budget...I try to spend as much money as possible with what I think will work best...I always test on Saturday and have a decent chance of breaking even by the end of the day. With a $2 offer I should have at worst 1-3 conversions by $20-$25...With a $10-$15 offer I should have something by the $60-$70 mark. Note that I'm starting with at least some traffic that has been profitable on other offers or in other countries and I'm also blacklisting "marginal" traffic. No reason to spend money on marginal traffic if it doesn't work on good traffic.
@kepe95, @sihlous
Thanks for the reply and explanation!
I guess i haven't really tested enough mobile campaigns and gained enough experience.
If a payout less than $3, after spending $50 with 30% CTR (on popular landers from spytool), ROI hit -70%.
Would you continue to test this offer? or usually for initial testing, what type of ROI must be achieved before you will willingly to optimize that offer?
@socceralien - it would depend on what the rest of the stats looked like.
Is that -70% across the board, or were there specific placements, OSes or carriers that did better or worse? Also, how did the banners do, and what was the CTR / CVR variance on the landers?
If the performance was evenly 70% with no peaks or troughs in any of those elements, I'd probably kill the campaign and move on. But if there's potential in some places, I might try to explore that, or might just note it down and move on to something new, depending on the potential volume from the more profitable spots.
Overall, my usual experience is that people are too slow to kill campaigns rather than too quick. Most people have realised that you need enough data to make a decision at this point - I don't see too many people giving up on an offer after 1x payout any more. But a lot of the time, folk will carry on trying to optimise banners and landers on a campaign when they'd be better off killing it and starting over with an entirely new angle, offer, traffic source and/or geo.
It just depends.
1) Traffic Source (impressions in this geo, vertical, keyword)
2) Audience Size (are we talking millions, how broad is the audience)
3) Number of Creatives (you should always be testing multiple creatives)
4) Experience with the vertical / offer type (this gives you some instinct)
5) Information provided from your AM / Advertising Partner (epc, cr, demos etc).
So with all of the above in consideration it really just depends on how familiar you are with the product / service you are promoting. If you have been running thousands of (joins, sales, leads, downloads) to this audience/vertical for months upon months and you are simply testing a new offer you should not have to go super deep with your testing budget. In this case you might rely on your AM or Advertiser to provide some basic data on (CR% or EPC) to better determine your testing budget. However, if you are trying something completely foreign to you in terms of what you are promoting, you might need to spend a bit more to gather some proper data and begin the process of trimming the fat. Obviously, like networks and advertisers, not all traffic sources are created equal. Do your homework and make sure the traffic source matches the audience you're targeting.
I hope this gives some additional insight to your original question.
Home > Paid Traffic Sources > Mobile