So i had an idea for PPV a while ago that seemed too good to be true - it offers endless scalability and LP's only take a few minutes to build. I've tried it a couple times in the past but only managed to break even at best. The problem is, the angle seems like it "can't fail" on paper, but actually getting it to work is easier said than done!
I'm posting it here incase anyone has any ideas to further it, or executes it differently and manages to get it working, because once somebody figures a working technique, the scalability factor is endless.
The Idea:
Offer the visitor to your chosen ppv targets (online stores) a coupon code for their purchase. But only choose targets that are at checkout stages, for example - domain.com/checkout.php would load your popup, but domain.com wouldn't.
This not only makes sure the visitor is actually buying something rather than just browsing, but also in most cases will enable you to bid just 0.01, escaping the competition of the root domain.
Here's a lander i've tried in the past for this idea:
Step 1):

Step 2):

Step 3):

I've got it set up so step 1 automatically redirects to step 2 (which is just for show, and displays a loading graphic for 5 seconds giving the user the illusion something is happening) then finally hitting step 3 which tells them the discount has been automatically applied to their order - if the user doesn't engage with the lander within one minute, the page is set to auto redirect to the next step, which eventually end up following the affiliate link through automatically.
The problem with this technique however can screw up tracking 202's CTR stats as i have step 1 > step 2 configured as a click-through but figure it's better to have this included in the process. So if you watch spy view whilst traffic is coming in, a good 80% will click through to the offer, but i'm unsure how many are being automatically redirected and how many are manually clicking through. Probably worth a split test to find out.
Another idea to attract attention would be to add auto-play audio, or some type of discount related youtube clip.
Using legit coupon codes will result in better conversions, but fake coupons have also been proven to work, the main goal is to get the current user to click your affiliate link but continue in the same window/browser as the popup loaded and continue with their purchase - from experience, if the coupon fails to work for the visitor, they often still go ahead and purchase anyway.
Final thoughts:
I think this angle is better to be cloaked if you manage to get it working on a big scale as most of the time you'll be "stealing" what would have been a commision-free customer to the advertiser.
I think the main reason why i've failed with this idea in the past is cookie related. If somebody is already browsing a site, and are at the stages of checkout - they will have a tab/window already in use, and as this PPV lander will load in a popunder (seperate browser window) if the user continues browsing in their old tab after following this process through, i'm pretty sure (but not 100%) that their old cookie will take over again, meaning we would not get a commission if they purchased.
Ideally, it would work best if when they user clicks 'continue with your order link' on step 3, it would close their browser, and re-open it with the affiliate link, and to take this even further, deep link back to the checkout page where the popunder originally loaded. But i'm not sure if that's possible?
Any ideas/feedback/tests welcome to further this idea! This one could be used on thousands of targets, so i can't imagine competition or saturation being a problem. 
You da bawse Lee!
ive done this in the past, it works, and ive profited, the only downfalls were not enough traffic to scale it up, and the other downfall is that certain traffic sources shut it down on me because they dont want me targeting checkout pages, they said it was against the rules and killed my campaigns
The problem is that it does not look "official" or believable. Everybody assumes that "ugly" and amateur landers work because everybody says they does. Not always the case.
Slap an Amazon.com logo on the top of those pages and you'll be rich (and I'm not kidding)
Obviously there are trademark issues and it is dodgy but i'm sure you will figure out how to do it legit.
I think adding the amazon logo would be against terms, but could be added in after approval (pretty risky). Bidding on checkout pages so far on trafficvance hasn't been an issue (15 or so campaigns i've tried so far have all been approved first time).
I've tried LI but the campaign gets rejected so i think checkout pages are against their terms.
The biggest issue i've had is getting volume like gts6 mentioned, but i didn't know you had to bid higher than the root domain for a deep target, so this is definitely something i'll try.
Has anyone else tried this method or something similar and had any success? I'm sure it has some life in it but needs to be tweaked to get it working more effectively first.
Another idea i had was simply redirecting the user through the affiliate link automatically when they land on a checkout target and deep linking them back to the checkout page, but i'm unsure what will happen if they don't continue shopping through the popunder window, as their old cookie could contain a different affiliate link in their previous window...
@gts6 - did traffic vance reject your campaign after it had been live for a while first? Or did this happen on another ppv network?
I know this sounds noob, but I like that styling on your pages.
All my LPs look harsh/html3 ish still to this day. I have a few icons sets that look like yours, but what font/css properties are you using?
@gts6 - when you say it didn't generate any volume, was you bidding higher than the root domain's bid? I've never tried to outbid the root domain when bidding on a deep target but i've read thats what you have to do to get decent traffic.
@eliquid - the css for the font is - font-family: "Trebuchet MS", Trebuchet, Arial, sans-serif;