How I cut bad placements, is it reasonable？ (4)
I'm glad you're finding the calculator spreadsheet useful. 
You've brought up some very good points. Here are my personal thoughts:
1)The spreadsheet is just a tool with limited capabilities, whereas we are humans capable of considering additional factors to fine-tune our decisions. Your additional cutting criteria are a perfect example of this. I also have similar criteria of my own that I use to cut placements before the calculator identifies them, but didn't want to complicate that calculator spreadsheet thread for the newer stackers. The calculator will catch those eventually anyways. But your criteria will save you some money.
2)"Impression >500 and 2. eCPC>0.05$" <--- I think those are fair. It would look like that your average CPM is around the same as your payout. This means you basically need 1 conversion per 1000 impressions to break even, so 500 impressions without a click I think you can safely cut. As for eCPC>0.05, it would highly depend on the average CR of your offer. Some super hot offers may convert 1 in 3 clicks - if that was the case then maybe you should set your cut-off eCPC value higher. But really, with 10k+ placements, you don't need to be too safe in your cutting approach.
3)If you're getting traffic from so many placements, and the amount of traffic isn't a problem for you, you can cut placements more aggressively and still have a lot of traffic after you're done. This means you can set the "confidence" level to a smaller number (try 70% - or even 60% which I never tried but if you want to speed up the testing at the risk of cutting out some good placements then maybe it's worth a try!)
4)"Can I add them to blacklist in other GEO? I don't know if they have similar performance in different countries." <--- Although I DO do this personally - I don't have enough experience to comment on whether or not this is a good approach - hopefully some of the pros can chime in. The performance of a placement can vary from geo to geo and from offer to offer. And then there are apps that tend to perform badly for almost all offers in all geos (e.g. flashlight apps) due to the nature of the apps. However, with so many placements at your disposal, I personally would just blacklist all the ones that are not sending major traffic, and maybe just re-test the ones that are. Or, just simply blacklist everything that were blacklisted for another geo (which like I mentioned is what I do).
Speaking of blacklists - just in case you're not aware, Mr. G got his hands on the most-often blacklisted placements from Decisive, which he shared here:
http://stmforum.com/forum/showthread...sive-Blacklist
What I personally do, even before I run traffic, is apply these decisive lists, plus blacklist grindr and all flashlight apps. Again - may or may not be the best approach.
5)"I will make a simple landing page, cut all placements if the LP CTR is lower than 10%" <--- I would NEVER cut a lander based on CTR alone. Ultimately CTR alone will not make you money - it's CR that will. There are things you can do to your lander that will boost CTR but bring down CR overall, and that's not what you want. For example, you can make bold promises on your lander to get people to click through, but if the offer page doesn't jive with your promises people will bail (which will be reflected in your CR). Conversely, some landers may have really low CTR but a good CR that would make it profitable overall. An example is a lander that makes people jump through hoops to "qualify" them for the offer - so that almost every one of those that are willing to jump all the hoops will convert once they click through.
Thanks for your post! 
Amy
Amy can u talk more about how you cut placement, before the calculator, for advanced stacker ? Maybe a new part 3 thread ?
Hey Ryan!
I could touch on that maybe in a post in one of the original threads, but "rules of thumb" like the "extra criteria" we're talking about, are the darndest thing to try to explain. It kind of borders on the realm of "gut feelings" and will vary widely with each particular situation - the offer, the traffic source, the number of placements, the amount of traffic available, etc.
So the best I could do would be to introduce this as "food for thought" rather than suggestions on how to determine these "extra cutting criteria".
I should stop here before using up all the quotation marks I have in stock. 
Thanks for the suggestion though!
Amy
Thansk very much, I will read your detailed answer three times, very helpful
Home > The Newbie Zone > Questions and Answers