Alright, here's a quick update Campaign 2:
Still not profitable. I'm going to pull the stats for all 4 days this campaign has been running and try to do some more optimizations.
Before I optimize sources/targets, I want to take a look at other factors. This may give away our campaign, but to be honest, I think it's more important to try to help you guys optimize than it is to hide the offer.
So far, all the optimizations that I have done on this campaign have been based on data from the
traffic source. I've optimized the sources and targets that
Zeropark is giving me traffic from.
I have not done any optimizations based on the visitor data.
Typical visitor data that you are going to want to optimize are:
- Devices
- OS
- Browser
- ISP
- Day parting
We're only targeting desktop computer with this campaign and a specific OS. However, with Zeropark, you can also optimize their redirect/pop campaigns based on the OS version. So, let's go ahead and take a look at that data:
There is some really good data here, but unfortunately, because of how Zeropark is setup for OS optimizations, it won't do us much good. When you optimize based on OS, you can select a MINIMUM OS version to target. In our case, we would want to EXCLUDE the 2 most recent versions of MacOS (10.14 and 10.13).
We can remove everything less than 10.9 because none of those versions have gotten us a conversion, but we're really not going to save much and we don't have enough data on those versions yet.
Something we could test is setting up a rule in
Voluum so that traffic from 10.14 and 10.13 is redirected to another offer and see if we can get that traffic to convert with a different offer.
For now, I'm going to leave it alone. Next, let's look at browsers:
My campaign is setup to target only Chrome and Safari because that's what the offer allows, but we have a clear winner when looking at our data here.
Our EPV on Chrome is $0.003 whereas Safari is $0.01915. We're seeing MUCH BETTER results with Safari so I will remove Chrome.
We cannot optimize this traffic on ISP so let's take a look at our day parting data:
There are a few hours of the day performing better than others, but really we do not have enough data here to make any decisions. We've spent $49.10, but most for most hours we only have $2-3 in data. The offer pays $2 so what we have isn't significant. What I will be looking for over the next week is blocks of hours of the day that are underperforming. Typically, you'll see this between like 1AM - 7AM. Once I see that, I will have Zeropark pause traffic during those hours.
We've made one significant change to our campaign by removing Chrome traffic. I believe as long as we continue to see similar results as we've seen over the past few days, the change to all Safari traffic will get us profitable.
Let's take a look at our sources/targets. I will likely not make many changes to these so I can see how the Safari change impacts everything.
Our source breakdown tells us that the only thing new is that we should pause gridelin-bear. It's now at 2x the payout and no conversions. The other sources are not profitable, but we're making progress.
When I pull the target report for badious-buzzard on Zeropark, I see we have some potential:
The top target is slightly ROI negative, but hopefully our Safari change will improve that. We've bought about 32% of the traffic available with our bid position showing 3 so I don't think I need to change the bid right now. We'll see how that changes over the next few days.
We've bought about 36% of the traffic on the second target and that one is very profitable. We don't want try to get the top bid on it yet because based on our current data that would put us ROI negative. Let's just leave it alone for right now and we'll see how the Safari change impacts it.
We do not have enough data on the other targets for this source yet.
coquelicot-bear is our ROI positive source so let's dig into it's targets.
First I sort it by Spent so I can make sure there's none I've spent more than $2 (the payout of the offer) on without a conversion:
That looks fine so let's now click Conv so we can see the targets we have conversions on for this source:
We have 4 different targets we have conversions on for coquelicot-bear. The first one shows 0 available so let's leave that alone (not sure why this happens sometimes). The second, we've bought 2 clicks and have 1 conversion. Really nice ROI! lol Our bid is too low for this target since we've only been able to buy 2/25 clicks so let's bump that up. I set a manual bid of $0.10 for this target and hopefully we'll get some more of those clicks.
The next 2 targets hardly have any traffic so we'll leave those alone for now. We're bidding enough on both of them since we're getting most of the clicks available.
Lastly, I'll do the same thing for stammel-snipe as I did for the previous sources. These targets seem to have a lot more potential so I am getting aggressive with my manual bidding on them:
As you can see, optimization is a long process. It takes time. You have to analyze. This is why it's so frustrating when people post a screenshot of their 200 clicks and 2 conversions and ask what they did wrong
I believe this campaign will be profitable soon. I will likely begin testing this same offer in other allows GEOs next week
Onward and upward