A Successful Template for Launching New Mgid Campaigns

Campaign Management:

First few sets of Ids that hit that are driving the majority of the volume.
Downbid those main Ids with all the subids 10-20%. (0.9/0.8 Co-Eff) We want to test them, but not spend the next 3 days just testing that source.

The Day 1 goal is to get everything spending evenly at mgid. There are a ton of sources out there, with unique bid prices. When you start a new campaign, its hard to know how the response will be. You want to “try” to get all the widgets sending a similar amount of clicks per hour.

Converting widgets generally show signs of life early on. So you want to increase the probability of hitting widgets that will convert in your funnel.

Down bidding those high volume ids. This will lower your spot in that source. The good news is 10-20% isn’t enough to drastically change the performance. It will lower impressions, and allow you to test a wider range of widgets. Make sure this gets applied to all subids. Once the main id gets a bid adjustment. Any new id that hasn’t been tested that gets a click, will come in at that price.

When the campaign is new, it doesn’t always have the weight to get onto the best converting placements for that niche. So you don’t want to idle super hard on the first sets of widgets you hit. You still want to test them, but don’t overbid due to a default bid on the ad. just adjust the co-eff to get them to all spend at a similar rate per hour.


When widgets start converting:

Depending on the volume. If its sent 1-5 clicks and converted. Chances are it has more volume. We are just underbidding, as others found this to be a good widget as well. Give this 20-30% if the volume is small. If the widget is high volume but profitable, try 10%. Give the campaign a day, and see how the results are. If you are still profitable, give it another small bump.

Make sure to also look at the last time you adjusted the bid on converting widgets. If you did not get an increase in volume, try a 20% bid. Might be heavily underbidding, and need to pay more.

Note; If you are at a “very high reach” – this means you are getting the majority of the impressions, and can’t really get any more from it. If the reach drop, then give it a bid adjustment. Otherwise, let it run, that’s the max it can do.


When converting widgets become unprofitable:

Take a look at the avg CPC you are bidding on that placement. You know at that current rate (bid) you get those results. If its -10 dollars profit most days but converting. Down bid it 10%, see if it becomes profitable at the new lower bid.

If the widget is sending high volume, and loss is growing. You can try down bidding 30%. See if it turns around. Otherwise, if the CR% doesn’t keep up with the volume. Likely will have to pause it. There are other better widgets out there.

Make sure the ad CTR and reach is still decent as well. If the ctr and reach drops on the ads. This could cause widgets to start to under perform.


When should I turn off the widget:
Everyone has different theories on this one of course. If the ads are proven and working, then you can judge a widget harder. If the campaign is new, and still being optimized. Then you will want to collect some data to justify your decision.

If a widget has spent 80% of the CPA (offer payout), with not the best LP CTR. Pause it.

If a widget has spent 65%+ of the CPA with almost no clicks to the offer page. Likely safe to pause it. At the current rate, and the cost you are paying per click to the offer page. You would have to have a crazy high cr% to be profitable. If it’s not there, likely won’t ever be.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.