When should I use CPA bidding vs CPC/CPM (4)
I wonder when should I use CPA bidding. What is the advantage/disadvantage of this bidding method vs CPM/CPC and how to combine CPA with CPC/CPM effectively.
This article https://propellerads.com/blog/adv-ta...-notifications makes it so easy but I don't believe it's that easy. There must be many situations in which CPA is not working.
Also is there any case in which the CPM/CPC campaign can not make profit but then the campaign turns to green when we change the bid to CPA ? Maybe the secret method of CPA is smarter than our traditional optimization method.
There are different ways how or when to run CPA Goal campaigns so I can only tell about how I personally do it.
I try to explain as good as possible why I do it the way I do it so that it should be easier to understand 
At first I´d like to tell a bit how the CPA Goal campaigns work.
When you start a CPA Goal campaign they often use high bids, sometimes such high bids that it would be unrealisitc to be profitable when these bids would continue.
For example when you set a CPA Goal of $0.50 but some bids are like $0.60 CPC, it just can´t work.
But the goal of a CPA Goal campaign is not to be profitable right away, the first step is to check for potential of the campaign and the best way to do so is to test the campaign on high quality traffic and such traffic is mostly more expensive.
The result of that approach is that a even very good campaign can run on losses in the beginning before the algorithm adjust the bids.
As we all know, a picture is better than thousand words so here I tried to show a bit how a CPA Goal campaign works over a longer timeframe
Of course that graph is not accurate at all and very rough but it shows pretty good the process of a CPA Goal campaign.
It starts with high quality and high bids to check for potential, then first conversions come in.
The algorithm adjust the bids and more conversions come in.
The platform then adjust the bids to set lower CPC to not that good converting placements and keeps higher bids on better placements.
At the same time it excludes non converting placements.
So when a good converting CPA Goal campaign runs longer the bids become lower and at the same time the conversions increase.
At one point it will run pretty stable then and doesn´t change that much anymore.
One thing that is always the same with all CPA Goal campaigns is that they start with high bids, the difference is if it converts good enough to be worth it or not.
And here we come to the reason why I don´t use CPA Goal campaigns to test offers.
For the Algorithm it doesn´t make a difference if you run a proven campaign or a completely new test campaign on CPA Goal.
The platform just can´t know it and treats all campaigns equally.
With a proven campaign I have no problem to eat the losses in the beginning because I know that the campaign is converting good so that it will be worth it.
When you test campaigns on CPA Goal, maybe even with different offers and/or landers to splittest, you don´t know if the campaign is a winner or a loser but you still have to pay the high bids in the beginning although you can´t be sure if the campaign will convert good enough or not to make good for the losses.
That´s why I test new campaigns first on CPM (pops) or CPC (push).
Doing so I can use my own stats and bids to check for potential without the risk to overspend on non performing campaigns.
When I see that the campaign is converting good I also set it up as CPA Goal and run it together with the other campaigns.
That way I can get most of both worlds.
Here is one of my campaigns that shows that CPA Goal campaigns can bring nice additional income without hard work
The campaign started more than 2 months ago and is still running.
The payout is €0.75, that´s why the $ payouts are always a little bit different based on the daily € -> $ exchange rate.
I set the CPA Goal rather low at $0.40 but it´s working very good that way.
The campaign is for push traffic, targeting only high activity.
As you can see, It started with lower ROI in the first days, then it got better and ran more stable for some time.
After a while the performance went down a bit for few days, maybe because the algorithm did some tests again.
That would also explain the higher volume.
Then the volume decreased for few days and since then it runs pretty stable.
Overall profit is about $400 in the tracker but the real profit is a bit more because I always set a higher % clickloss in my tracker.
This is only one campaign for one offer in one geo with one activity level.
I have few more campaigns running in the same geo for the same funnel but with other activity levels and also test different bids so even when one campaign doesn´t make that much profit when you run several of these campaigns it can be really worth it, especially when you run good converting offers with higher payouts than $0.90
Amazing knowledge @twinaxe. Thank you
I have a question: When I bring a CPM campaign to CPA Goal, should I also bring along the Blacklist which I accumulated during the time I run CPM campaign ?
@twinaxe: I tried to move one POP campaign to a CPA campaign today and the performance is embarrassing as can be seen in the figure below:
The green one is the POP campaign with payout 1.3$. I ran it for about 2-3 days. I copied the BL from the POP campaign to the CPA campaign.
I have some questions below:
1. Do you set budget for the CPA campaign (like standard 10xpayout). To which point should I stop the CPA campaign to avoid burning budget like this one. Initially I kept the CPA campaign running longer in the hope of achieving the stable point but it quickly spend 66$ in about 12 hours. I think this is too much spending so I stopped it for now.
2. Do you have any minimum cut-off parameters for a green CPM campaign before launching a CPA campaign: like the number of running days, the min ROI, min number of conversions. I think in order for a CPA campaign to work, the CPM must be stable for a certain period right ? In my case do you think I move to CPA campaign to soon because my CPM campaign is only at a break-even point for few days ?
Home > The Newbie Zone > Questions and Answers