PPV Testing Questions

by mream
10 replies
I have been trying for a long time to make some money with CPA. A long time. I realized one of my problems is not being rigorous enough about my testing. Now I'm trying to do better. I have plans to follow including the amount of impressions I want to see, how much I'm spending, etc. I'm trying it on POF and some PPV sites and I'm hoping this approach will work better.

With PPV, though, I have a question. I see two ways to go about the testing. One way is to throw up as many URLs (or keywords, but I'm doing URLs now) as I can and see if anything converts.

The other way is to start with a small list, let them get the number of impressions I'm looking for, then add more and stop the ones not converting.

I'm not sure what is better. It seems obvious that the first method will be more expensive and the second will take longer. Right now, I have more time than money so I'm going slow. Maybe it's too slow?

My concern with method 1 is that I'll blow through a bunch of money, get a lot of TOTAL impressions, but not a significant amount to any single URL to determine if it's a keeper or not.

Am I overthinking this?

Thanks,

Matt
#ppv #questions
  • Profile picture of the author davidmeeonline
    From my experience, if you do good research before starting anything, you can usually get some conversions and go from there.

    To start simple, say a free download, I check the website on quantcast to get an idea of demographics, and spyfu to get some keyword ideas. I then decide which is the best, buying keyword i have and go to CPV den and use the url scraper to get 100 urls which are very targeted.

    Put them into your PPV network and track them through Tracking202 or Prosper202. If you get some conversions, you can then use those urls and get more targeted urls based on your converting urls.

    If none convert you could try another 100 keywords until you get some to convert.

    Anything more involved than email submits and downloads I believe need a good landing page to get converting.

    Hope that helps

    Dave
    {{ DiscussionBoard.errors[2498107].message }}
    • Profile picture of the author mream
      Great tip, Dave. Thanks.
      {{ DiscussionBoard.errors[2498879].message }}
  • Profile picture of the author harro1
    Originally Posted by mream View Post

    I have been trying for a long time to make some money with CPA. A long time. I realized one of my problems is not being rigorous enough about my testing. Now I'm trying to do better. I have plans to follow including the amount of impressions I want to see, how much I'm spending, etc. I'm trying it on POF and some PPV sites and I'm hoping this approach will work better.

    With PPV, though, I have a question. I see two ways to go about the testing. One way is to throw up as many URLs (or keywords, but I'm doing URLs now) as I can and see if anything converts.

    The other way is to start with a small list, let them get the number of impressions I'm looking for, then add more and stop the ones not converting.

    I'm not sure what is better. It seems obvious that the first method will be more expensive and the second will take longer. Right now, I have more time than money so I'm going slow. Maybe it's too slow?

    My concern with method 1 is that I'll blow through a bunch of money, get a lot of TOTAL impressions, but not a significant amount to any single URL to determine if it's a keeper or not.

    Am I overthinking this?

    Thanks,

    Matt
    I would suggest you to start with second method which you mentioned this way you will not lose much money, add 50 url's in starting, use quantcast or alexa to get demographics, remove url's which are not converting then make another search on quantcast and find another 50 url's, rinse and repeat. This method has been somewhat profitable for me so far.
    {{ DiscussionBoard.errors[2499249].message }}
    • Profile picture of the author mream
      thanks Harro1. That's kind of what I've been doing. How many clicks to you consider a valid test?
      {{ DiscussionBoard.errors[2499423].message }}
      • Profile picture of the author EdwardDennis
        Originally Posted by mream View Post

        thanks Harro1. That's kind of what I've been doing. How many clicks to you consider a valid test?
        Hi,

        it depends on the cost per view of that URL. But, my rule of thumb is 100 clicks. Sometimes I multiply the views and the cost per view. If the total cost hasn't exceeded the payout, I would keep the URL/Keyword active. Hope that helps.

        Best of luck,

        Edward
        {{ DiscussionBoard.errors[2500395].message }}
      • Profile picture of the author davidmeeonline
        Originally Posted by mream View Post

        thanks Harro1. That's kind of what I've been doing. How many clicks to you consider a valid test?
        Edward is right, it does depend on the cost per view, this is the area where everyone has a different method they use. I would wait for 200 views before looking closer at the url, and wouldn't let it go above 800 or so if it had no conversions. I would only let it get that hight if I had a large payout for that offer.
        Small payouts I cut off a lot quicker, as it wouldn't be worth it letting it get up above 300 views with no conversions.

        Cheers
        {{ DiscussionBoard.errors[2500901].message }}
  • Profile picture of the author Per Action Rob
    I think for most novices you should start with one solid target say QVC.com for example and rotate the **** out of offers to that one target and see what you have.

    I've spent close to 400 grand on CPV and I can tell you with fair certainty that if you are willing to burn some cash and stick it out, you can generally find not one, but a few offers that will do well to a specific target.

    It's also a lot more controllable, it's one god damn target. I used to employ the same strategy with PPC, I'd get a high CTR ad rolling first and then rotate the destination for offers in that category.

    Insanely simple, very effective, most of marketing works that way.
    Signature

    See thousands of dollars in live affiliate marketing media buys each month $1 30 day trial:

    http://www.MediaBuyingMonthly.com

    {{ DiscussionBoard.errors[2500941].message }}
  • Profile picture of the author KC-Coop
    As far as how much to spend per URL - I typically spend between 1-1.5 times the payout and if it doesn't convert I pause it.

    If you have a low budget - I would take it slow and only add higher volume targets - and not too many at once.
    {{ DiscussionBoard.errors[2501063].message }}
  • Profile picture of the author WebPen
    Yeah pick one high volume site and start rotating offers- that way you really know the demographics and will probably lose less money before you figure out what works.

    Testing 3K sites, even if each only gives you a few impressions a day, will eat up budgets pretty darn fast.
    {{ DiscussionBoard.errors[2600552].message }}
  • Profile picture of the author Firstrate
    Consider looking at CPVLab.com for your tracking. It's just come out and is setting the industry alight right now.

    Amazing tracking and optimization capabilities.
    Signature

    {{ DiscussionBoard.errors[2605410].message }}

Trending Topics