Day 11: Introduction to Testing & Optimization (Voluum) (6)
Day 11: Introduction to Testing & Optimization (Voluum)
In the previous lesson, we've set up your very first campaign together.
The $10 budget should have been spent by now. So we should have some stats to look at!
You may have made your first conversion by this point - if that's the case, then CONGRATULATIONS! Seeing your first conversion is always exciting! And I wish you many, many more in the near future.
If you don't yet see your first conversion, don't be disappointed - they will come, and very soon.
I'm not going into details on HOW to optimize a campaign - because we likely don't have enough conversions at this point to make any optimization decisions.
What I'll do in this lesson is just to guide you through how to drill down to various stats in
I'll talk more on optimization in later lessons, but this primer will prepare you well for what's to come.
***********************************
ACTION & EXPLANATION
In this section I'll be weaving explanation into action, as I'll be asking you to drill down into various stats, and describing briefly what we're looking at/for. After that I'll have a separate EXPLANATION section to cover further explanations.
In
Click on your campaign's row to select it, then right-click to bring up the menu and select "Report".
A new "tab" will open up within Voluum to display the stats for that specific campaign.
You'll see 3 drop-down menus:
In Voluum you can drill down to a maximum of 3 levels deep.
Click on the first dropdown and you'll see lots of different variables you can display data for:
If you scroll down this list of variables, you'll find a bunch near the bottom that start with "V#:". These are values that are passed back by the traffic source - which in this case is PropellerAds.
So where did these names come from? Remember when you were adding PropellerAds as a new traffic source in Voluum, you loaded a template of PropellerAds that was provided by Voluum? Let's take a look right now.
Click on the "HOME" tab:
Click on the "Traffic sources" tab, click on the "PropellerAds" row to select it, then right-click and select "Edit traffic source" and you'll see in the new window a list of tracking tokens. In the "name" column is where the variable names are defined:
And how does PropellerAds pass these values? Remember when you were creating a campaign in Voluum, you had to choose PropellerAds as the traffic source? When you did that, the resulting Voluum campaign url automatically included PropellerAd's tokens:
Then what are all the variables above PropellerAds' "V#:" variables? Those are Voluum variables - values that are detected by Voluum. Voluum too has the ability to detect information about each visitor sent by the traffic source, such as what device and OS they're using etc. etc. - just like PropellerAds can.
So all these variables you can choose in tracker reports are a mix of values detected by Voluum and values passed back by PropellerAds every time it shows a visitor your campaign url.
(Don't worry too much if you're not understanding how tracking tokens work - I'll explain tracking in detail in a future lesson.)
As was mentioned above, you can drill down up to 3 levels deep. For example, try drilling down to "Device types" -> "OS" -> "Browsers", like so (don't forget to click "Apply"):
Spend 15 minutes drilling down to the various variables and different levels of different combinations yourself, just to get an idea on what kinds of data is available. All this data can give you the insight required to make wise campaign optimization decisions. This is one of the main reasons why having a tracker is so important.
***********************************
EXPLANATION
First of all, let's define the goal of optimization.
And how can you accomplish that goal? Answer: By controlling the 3 main campaign elements:
And how do you optimize a campaign to profits by manipulating these 3 campaign elements?
2)By controlling which traffic segments to target - which can be achieved by a)optimizing the bid, and b)optimizing campaign targeting (country, device, browser, carrier, etc.) at the traffic source.
(If you're not familiar with how pop ads work: The traffic network - PropellerAds in this case - are in collaboration with webmasters that are willing to add to their site a piece of code given by PropellerAds, that would allow PropellerAds to display ads on their site. I refer to these sites as "placements" - they're also called "publisher sites" by the traffic source. PropellerAds charges us to show our ads on these publisher sites, then pays a part of that money to the webmasters.)
In any specific campaign, and at any given time, when you drill down to the various variables to view the different traffic segments, some will be in green (i.e. profitable) and others will be in red (i.e. in loss). The trick is to either just target the profitable segments, or to cut the unprofitable segments to make the campaign profitable.
Everyone does testing and optimization a bit differently, but here's the general approach in a nutshell:
2)At the traffic source, either a)target the profitable traffic segments (an approach also referred to as "whitelisting") or b)exclude the unprofitable traffic segments (an approach also referred to as "blacklisting"). (The testing of bids is also a necessary part of this process - more on this in a future lesson.)
***********************************
Of course there is a LOT more to campaign optimization than just what I've described above, and I will go into campaign optimization in more detail in the next lesson and in future lessons. But I hope this lesson has provided you with a bird's-eye-view of what campaign optimization involves.
Amy
Hi vortex, I would like to know if you will continue to update this guide?
Originally Posted by tracyw020
Thank you for your support!
Amy
Hi Amy, first of all thank you very much for your post and the hard work it is behind. Really appreciate it!! I´m new in CPA marketing and is so good to see detailed guides like yours!
I have a question Amy: Why you should make, lets say for example 2 differents campaigns of the same offer that are different each other because one is targeting desktop devices and the other one targets smartphone devices. I mean, you can make a campaign that focuses both devices and 
but as I said I´m just a noob haha.
Originally Posted by valenciacpa
I have a question Amy: Why you should make, lets say for example 2 differents campaigns of the same offer that are different each other because one is targeting desktop devices and the other one targets smartphone devices. I mean, you can make a campaign that focuses both devices and
but as I said I´m just a noob haha.It's a great question - there are no stupid questions.
My thoughts are that you CAN target both in the same campaign. The main concern here is that some trackers (e.g.
So if you're targeting desktop and mobile in the same campaign as well, that would be yet another level of data.
Therefore, for the sake of being able to drill down into more levels of meaningful stats to make informed campaign optimization decisions, it would be good to eliminate as many variables as it is practical. For example, by targeting only one country and one device type (either desktop or mobile) and only one connection type (either wifi or carrier) in a campaign.
Another good reason for targeting more narrow (only one country, one device type etc.) is so you can bid more appropriately. Take countries for example - each country has different average bid prices, so if you're targeting multiple countries in the same campaign, you can only assign one bid, which may be too high for some countries and too low for others. The same thing applies to desktop vs. mobile, and wifi vs. carrier. In most cases it would be better to test bids for each separately.
But I'm speaking in a general sense. There are of course exceptions. For example, for initial testing I would often test multiple countries at the same time just to see which country shows promise. But then I would start a separate campaign for each country that shows promise so I could bid appropriately and optimize further.
There are a million ways to test and optimize a campaign. Everyone will develop their own style eventually. I encourage you to use the 40-day tutorial as a starting point only, and do things your own way - which may be different from how I do it.
Hope that answers your question!
Amy
Thanks for the fast reply!!
Ohh, I see now. Makes lot of sense. I started yesterday my first campaign and I was kind of struggling with the bid price hahaha. Like, how do you know which is the correct bid? because of the estimator that appears? (my traffic source was mgid) is it reliable? I should put the average bid that it appears, a little higher , lower?
For how long should I run the campaign? Hw many days is the average to test if a campaign works or not? And the budget of the campaign I read in the forum that should be 3x they payout of the offer? So in this case the payout is 20 $. I should try this campaign with a total budget of 60 $? Is like that?
Again, thank you so much Amy!! There is not that much information of CPA, like extensive guides, videos of the whole process etc specially in my mother tongue: spanish so it´s so nice to be part of this great forum and incredible community!
Best regards,
Jaime
Home > >