I'm in the optimization stage of a promising MGID campaign, and I've started adjusting my bids on individual widgets based on EPC. So, I'm using the coefficients in MGID to increase (or decrease, if necessary) the widget bids to ensure I'm getting traffic to widgets that are proven to convert.
I'm trying to figure out an optimal percentage of EPC to set my bids to on a per widget basis. Right now I'm setting them for approximately 75% of EPC.
When you find a widget that is proven to convert, what percentage of EPC would you begin testing with? I'm concerned that, since so many widgets spend money without converting at all, 75% might be too low for those that do.
Of course raising a bid past the point where traffic volume is maximized is pointless, so I'm also curious about how experienced MGID users find the optimal bid on a per widget basis.
Ultimately this completely depends on your ROI objectives, however I do think that bidding 75% of a widget's EPC is way to high.
You are basically setting yourself up for a projected 25% ROI on converting widgets only, however for every profitable widget you will have dozens or even hundreds of widgets which are loosing money which will make it extremely difficult to make the campaign profitable as a whole in my opinion. Unless you are running a whitelist campaign, and even then i would say it's to high.
I am usually greedy with performing widgets and try to get the highest ROI from them as possible so that after my spend on non performing widgets the overall campaign ROI evens out.
A more cost efficient way to get more traffic from these widgets would be having a better flow, especially higher ad CTR
Thanks Daanja. I posted this, because as I found more and more converting widgets, I started to realize that I was creating a situation like you described with a small ROI on converting widgets mixed with hundreds of non-converting widgets.
So when a widget converts after running a campaign and has had two clicks or twenty clicks to it, it's clear that you need to bid higher for that widget. What I'm looking for is a systematic approach to get more traffic to converting widgets without having to raise the bid on everything. I like the idea of bidding x% of ROI, because I can set this in a spreadsheet and quickly upload different bids for tens or even hundreds of widgets. But maybe I'm looking at this problem the wrong way.
I like your approach, as a matter of fact I do it very similarly when running campaigns on mgid.
I think you are already developing a very systematic approach that makes a lot of sense, bidding what is justifiable in terms of EPC on the campaign, while adjusting bids per widgets based on EPC.
Just keep tweaking it according to what makes sense according to your objectives while gathering more data and you should come up with your "secret formula".
Just make sure not to overlook your flow and ads CTR, which can make your system work even better by increasing the EPC and reduce click costs