When should I Stop Running Ads?
Optimization process is the next step after you?ve launched your campaign. It might seem very complicated to choose the moment for cutting ads that just don?t work right. Well, there are some issues, but it won?t be extremely hard.
There is one common mistake that affiliates make when trying to optimize their campaign. They don?t consider the number of views and look only on the conversion rate. Then they reject ads that don?t have the target rate at certain point (e.g. 0.5%). It?s a very bad idea. The thing is that such ads might have zero CVR when have 1000 views and 2% when it was viewed 4000 times.
How to use statistical methods to make the right choice?
If we say it in plain words, Statistical Significance shows you the number of specific events you need to be sure you?ll get the same result each time. That principal will help you to calculate minimal CVR for ad to be profitable and minimum CTR.
Before you reject any ad, define the conversion rate level you need to get a decent profit on it. The things you need to calculate this are your target ROI, CPM and the level of payouts from conversion. In case your source of traffic uses cost per click, there is a formula to transform it into CPM. The formula is the following:
CPM = CPC*CTR*10
To make it more clear, lets presume your CPC is 0.2$ and the result CTR is 0.2%. Using the mentioned formula we?ll get 0,4$ CPM.
There is also one important thing about all this. You should do the statistical testing only after you?ve rejected bad placements. Don?t use data with a lot of impressions from bad placements.
Saving money when you have small budget
If your funds are limited, saving money is essential. The one way to do that is to cut ads using their CTR as a measure. In case the last ratio is very low, while your CPC is very high and the CVR to make a profit is incredibly big, then be sure to reject ads with maximum predicted Click-through rate below the point.
There are two steps needed to do that. The first action is CPM ($)/payout ($). The second one is to divide the quotient by 5 (if we take 50% CVR level). The result is the lowest possible CTR to be above break-even point. Using the mentioned actions, you can also calculate your maximum expected Click-through rate. Just change the amount of conversions on the number of clicks.
Again, if the resulting ratio is below profitable CTR, you can cut the ad. Note: run the test only when the ad spends from 1x to 5x your payout. In case it does more, then testing on CVR rather than on CTR should be started.
Choosing a relevant ROI level
When you invest your funds in something, you expect to receive profit in return. ROI is a number, which shows you the profit you get on all the invested money in percentages. If it is more than zero, then ROI is above break-even point. But is it really enough to have ROI level of few percentages? Lets presume your daily spending is 500$. In this case a 5% ROI will award you with 25$ the same day.
On the other hand, you?ll probably consider 200% ROI as very tempting. However, there are some obstacles you?ll meet trying to reach that. First, you?ll have to reject many ads and invest a lot of extra funds in the campaign. Then, such high ROI is unlikely for broad campaigns. It?s more likely to reach this level in niche campaigns, still with great efforts.
If we take broad campaigns, ROI between 35 and 45% will be relevant. Remember that you?ll have a possibility to scale up. In case you?re running a niche campaign, 75 ? 85% is the level you should aim.