Optimizing PPV Campaigns - Basic Guidelines? (5)

These are basic questions but I felt they needed a post of their own and a little clarity, even if only for myself.

There are two main types of PPV Campaigns:

Direct Linking

Usually you optimize direct linking by testing a group of offers, find a converting offer, and then from that point remove all targets that don't produce a positive ROI direct linking because there isn't anything you can do to improve the conversion rate and there is no LP involved.


Campaigns with LPs

This is where I need a little clarity.

When using LPs in a campaign, initially you test two LPs at a time and rotate multiple offers.

Is it fair to say you can't really remove many targets until you've also found a converting offer? Unless a target is just driving alot of volume with very low CTR.

Then once you've found a converting offer, let it run until you've gathered enough data to remove certain targets that aren't atleast close to breaking even?


If this has already been discussed on the forum, you can point me to the thread, I looked around but didn't see any direct info to these.

Yeah I was feeling a bit the same about this yesterday and wrote about it in my follow-along. I feel like until i've really got this down I'm unable to juggle too many campaigns. As far as improving a Direct Linked offer you can further optimize it through Day/Week parting and only run the campain at times/days when it is profitable. I'm only testing one offer at time at the moment so I can target my LP more, plus you'll need a higher test budget if your rotating bioth LP's and offers which isn't easy on a budget! Then when finding a converting offer find the same offer on other networks to split test then. I'm still learning too so don't take that for gospel lol! I want to hear others on this too. I came up with the following as some guidelines yesterday, feedback appreciated! -

First Testing Round

1)Test 3x offer payout per variable ie 2LP's/1DL test budget will be 9x

2) If Campaign has no conversions after first test budget, kill it, unless one URL has used more than 2/3 of test budget. In which case, stop that URL and run test for a second round.

3) Kill DirectLink if it has no conversions after 3x offer payout

4) Kill any url's that have spent 3/4+ of offer payout without conversion

5) After first round of testing budget if there is "significant" difference in LP results, either split test new angle or start tweaking the winner, otherwise collect more data if they are getting conversions.

Second - Fourth Testing Round

1) If a URL had a conversion in the first round but no more, stop URL once it has spent 2x payout.

2) Further remove non-converting URLs that have reached their spend limit

3) Keep tweaking LP to try and improve CTR/CT and play king of the hill.

After Four/Five Rounds I will move any profitiable URL's into their own campaign and raise budget to scale it and further optimize with any day/week parting. I will then test any high affinity sites to the profitable target if I havent tested them already.

This is exactly how I test my campaigns on PPV - great post man!

Originally Posted by iwanturcoin

Yeah I was feeling a bit the same about this yesterday and wrote about it in my follow-along. I feel like until i've really got this down I'm unable to juggle too many campaigns. As far as improving a Direct Linked offer you can further optimize it through Day/Week parting and only run the campain at times/days when it is profitable. I'm only testing one offer at time at the moment so I can target my LP more, plus you'll need a higher test budget if your rotating bioth LP's and offers which isn't easy on a budget! Then when finding a converting offer find the same offer on other networks to split test then. I'm still learning too so don't take that for gospel lol! I want to hear others on this too. I came up with the following as some guidelines yesterday, feedback appreciated! -

First Testing Round

1)Test 3x offer payout per variable ie 2LP's/1DL test budget will be 9x

2) If Campaign has no conversions after first test budget, kill it, unless one URL has used more than 2/3 of test budget. In which case, stop that URL and run test for a second round.

3) Kill DirectLink if it has no conversions after 3x offer payout

4) Kill any url's that have spent 3/4+ of offer payout without conversion

5) After first round of testing budget if there is "significant" difference in LP results, either split test new angle or start tweaking the winner, otherwise collect more data if they are getting conversions.

Second - Fourth Testing Round

1) If a URL had a conversion in the first round but no more, stop URL once it has spent 2x payout.

2) Further remove non-converting URLs that have reached their spend limit

3) Keep tweaking LP to try and improve CTR/CT and play king of the hill.

After Four/Five Rounds I will move any profitiable URL's into their own campaign and raise budget to scale it and further optimize with any day/week parting. I will then test any high affinity sites to the profitable target if I havent tested them already.

Here is something I'd like to point out.. On the campaign I'm showing in my Follow Along (see link in signature).. I've testing about 25 or so offers.. 15 Victoria Secrets submits.. and then 10 different offers from an iPad type submit. I found 1 Victoria Secret offer that converted and so far I've found 2 iPad offers that convert fairly well.

So if you're using landing pages and you're only getting a 3-5% CTR and sending it to 3 offers that just don't convert.. you would soon enough determine that campaign was a failure.. cause you'd have to spend alot to get enough clicks to each offer to check if they convert. I have been using 20 clicks to each offer to determine if they convert or not. So 2200 views @ a 3% CTR = 66 clicks / 3 (offers) = 22 clicks to each offer... well 2200 views @ 0.015 = $33 that should be the minimum to spend per campaign test eh?

And like in my follow along, my CTR and LP metrics have been solid, my targets are extremely focused.. and I had previously paused a high volume target because it was eating up my budget and not receiving any conversions.. then when I finally find a converting target, I turn back on the high volume target I paused and it starts producing a steady amount of leads.

Just food for thought before ditching a campaign with no conversions.. if your metrics are strong keep testing offers to find one that actually converts before eliminating too many targets.

@canopus Cool, is that approach working well for you? ie are you profitable? don't mean to be overly direct just want to know if i'm on the right path

@ mr payne You have a good point the only thing is that CTR isn't directly related to CR, and it may not even be the offer that is not converting, it may just be that the LP doesn't convert for that target, I noticed it can really vary from group to group and even URL to URL within the group. I found this pretty interesting this week, i was running a zip submit as well across 5 targeted groups with the same 2 LP's and I got nearly inverse results between the LPs against the target groups and then even URLs against URLs. for eg LP1 was getting a CTR of >1% across 4 target groups and then about 5% in the 5th group BUT within that group that group it was only one URL that it was performing getting 8.5% CTR with 5 converisions and a CR of about 30% I know its not a huge sample but it just shows that what really hits the mark somewhere can be a dud with another URL that you thing is highly related. I also had in one URL LP1 with a CTR of 6% with 0 conversions and LP2 with 3.5% CTR with 3 conversion, so CTR was kinda irrelevant to conversions in that URL. The landing page was'nt highly targeted either was just a coupon style with a flashing Congratlations youve won a $1000 gift card thing, and the URL was gerber.com, a baby related site, and it performed poorly on the coupon related sites I targeteted, so u just never know.

I would be wary of testing that much on email submits tho cause even if you find a winner and spend $33ish/campaign if you spend $150 testing 5 campaigns, if not more, and find a winner I think there is a pretty good chance your converting offer will get scrubbed before you recoup all your testing costs. I was advised of this other day by several people and recommended that you're much better off testing your money in a more stable niche.