SeoDave's Guide to Generating High Converting Zip/Email Submits

Status
Not open for further replies.
Well to add to this big ass thread now, SEODAVE good idea and congrats on the profits, but one thing I would try out on your landing page is split testing your offers and monitoring clicks, it can be done easily with a database and a few lines of code, here I'll explain myself a little better.

You've got two buttons on your voting system, one is for YES and the other is for NO, instead of having them both point to the same offer have YES point to Offer A and have NO Point to offer B, now clock the links through Javascript or PHP (Depending on your hosting provider), now the user can't see the links in there status bar.

Ok if your doing a PHP re-direct type of clock, just before you redirect them increment a single value called clicks in a database table, then shoot the user out to the offer, it won't take longer then a few milliseconds to do.

Now you know how many clicks the YES button gets and you know how many clicks the NO button gets, and you've got web stats set up so you know how many impressions your getting, you also know how long people are staying on your site (Visit Length is good for re-working your designs to maximize stay times).

Anyway now you can easily split test two offers from the same landing page and you can figure out which converts more over a time frame, just set yourself a set amount of CTR before changing the offer, keep the better performing offer all the time and boom a bit more cash.

Everyone can even take this one step further with a radio button list poll, have the user select an option (1 out of 4 say), each different option brings them to a different offer, when they click submit, track the click (The radio button selected). Now you can track 4 offers at once instead of 2.

Tons of good info in this thread keep it coming. :)
 


Aequitas said:
Well to add to this big ass thread now, SEODAVE good idea and congrats on the profits, but one thing I would try out on your landing page is split testing your offers and monitoring clicks, it can be done easily with a database and a few lines of code, here I'll explain myself a little better.

You've got two buttons on your voting system, one is for YES and the other is for NO, instead of having them both point to the same offer have YES point to Offer A and have NO Point to offer B, now clock the links through Javascript or PHP (Depending on your hosting provider), now the user can't see the links in there status bar.

Ok if your doing a PHP re-direct type of clock, just before you redirect them increment a single value called clicks in a database table, then shoot the user out to the offer, it won't take longer then a few milliseconds to do.

Now you know how many clicks the YES button gets and you know how many clicks the NO button gets, and you've got web stats set up so you know how many impressions your getting, you also know how long people are staying on your site (Visit Length is good for re-working your designs to maximize stay times).

Anyway now you can easily split test two offers from the same landing page and you can figure out which converts more over a time frame, just set yourself a set amount of CTR before changing the offer, keep the better performing offer all the time and boom a bit more cash.

Everyone can even take this one step further with a radio button list poll, have the user select an option (1 out of 4 say), each different option brings them to a different offer, when they click submit, track the click (The radio button selected). Now you can track 4 offers at once instead of 2.

Tons of good info in this thread keep it coming. :)
Click to expand...

i would like to point out that by making the "yes" button go to a different offer than the "b" button, you are changing too many variables to determine anything consequential. people who click "yes" on a super divided issue like the mike vick case are certainly going to convert at different rates on different offers.

change one variable at a time, and test 'em against each other. you don't have to only test two, feel free to test a handful and go with the best. see which offers convert best for "yes" and which offers convert best for "no," and then use the respective offers.

good post seo dave, although it looks for sure like this idea is going to get swamppped now. like mkrongel said, execute it with your own twist and you should find success.
 
Dave,

Thanks for the post, ive attempted 2 campaigns now, 1 of which I seem to have butted heads with you on (PM me about that one :P )

Anyway, it may just be that I have a crappy Adwords account or something but the last 2 campaigns I did were met with immediate $5 min bids. For both of them im using 3-5 paragraphs of text with keywords sprinkled in. Any ideas?
 
lazyD i had the same problems, there was no buffer time to profit from it went straight to $6 bids in adwords...
 
anybody knows how can i make a poll or yes no voting and display its result with view result link.. i know its a php script..

i mean just making the user fool with a yes no button isnt enough i think.. if the user press back back buttons and want to check if the site was real or scam.. he might find the site actually fooled him to capture his emails zip details etc..


i have seen many php script and free ones too.. which offer voting or poll

but problem is those script redirects to result page after voting is done... we want it it to go to the offer pages ..
 
surf4fun said:
anybody knows how can i make a poll or yes no voting and display its result with view result link.. i know its a php script..

i mean just making the user fool with a yes no button isnt enough i think.. if the user press back back buttons and want to check if the site was real or scam.. he might find the site actually fooled him to capture his emails zip details etc..


i have seen many php script and free ones too.. which offer voting or poll

but problem is those script redirects to result page after voting is done... we want it it to go to the offer pages ..
Click to expand...

Just IFrame the offer page in the poll results page

M
 
Jesus... getting back to whomever posted on the steve fossett deal... I set up a little page last night... min bids were .10

I got 4 clicks before it ZOOOMED lol (only paid .34 for those 4 clicks though so I guess its not too bad.

With the breaking news stuff... You have to be right on top of the adwords portion to make sure your not gonne be paying the same price for a click as you would earn on a lead lol

Mehh... I took my meds after dinner and passed the fuck out.... now here I am.. all dissorientated... least I made it here safely lol
 
Hmmm..
How about randomizing the yes/no button?

Basically have a script (javascript or PHP) write one of the offers to one of the buttons by random.

Then track the stats.

Another ananlysis might be to leave the buttons pointing to one offer, but see wether YES-men or NO-ma'ams convert better.
Then screw around with the offer for the bad converting ones, or send them to another page or whatever.

::emp::
 
wsmith810 said:
Jesus... getting back to whomever posted on the steve fossett deal... I set up a little page last night... min bids were .10

I got 4 clicks before it ZOOOMED lol (only paid .34 for those 4 clicks though so I guess its not too bad.

With the breaking news stuff... You have to be right on top of the adwords portion to make sure your not gonne be paying the same price for a click as you would earn on a lead lol

Mehh... I took my meds after dinner and passed the fuck out.... now here I am.. all dissorientated... least I made it here safely lol
Click to expand...

I'm brand new to this. What makes the CPC shoot up like that? Is it just competition or something else? My two campaigns are related to breaking entertainment news and have only gone up to $.15. Does Google give preference to being first?

Also, is anyone using the content network or just search? I'm just getting a ton of empty impressions with no clicks from content, which is driving my CTR waaaaaay down. Does that have a big impact on my QS? Should I shut off the centent network?
 
SeoDave said:
Step 1: Pick a popular news topic (something big that will get a lot of searches is good)

Step 2: Scrape keywords from several of the top news stories online. When I've done this, I've actually grabbed quite a few news article titles as keywords.
Click to expand...

OK. I just got PavarottiFuneral(dot)com. Given that I'd like to make money from the site and it's a rather sensitive subject to any potential visitors, I'm faced with a dilemma: put an AdSense banner at the top *or* along the sidebar... with my luck he'll be cremated and buried privately :(
 
One of the things I love about COPEAC is the ability to use subid's to track conversions for different offers and landing pages. I send traffic to a page on my server that loads a random offer into a frame based on a list of predefined offers I have chosen and also inserts into the offer url subid field both the landing page info and any other info i want to track.

Here is how it can be used for the Vick poll example.

PHP: 
<?php

  if ($_REQUEST['uservote']) { $uservote = $_REQUEST['uservote']; }
  $landingpage = "vickpoll";

  $promourls[0] = "http://www.cpaclicks.com/redirect.asp?a=xxxx&b=xxxxx&d=0&l=0&o=$landingpage-$uservote&p=0";
  $promourls[1] = "http://www.cpaclicks.com/redirect.asp?a=xxxx&b=xxxxx&d=0&l=0&o=$landingpage-$uservote&p=0";
  $promourls[2] = "http://www.cpaclicks.com/redirect.asp?a=xxxx&b=xxxxx&d=0&l=0&o=$landingpage-$uservote&p=0";
  $promourls[3] = "http://www.cpaclicks.com/redirect.asp?a=xxxx&b=xxxxx&d=0&l=0&o=$landingpage-$uservote&p=0";
  $promourls[4] = "http://www.cpaclicks.com/redirect.asp?a=xxxx&b=xxxxx&d=0&l=0&o=$landingpage-$uservote&p=0";
  $promourls[5] = "http://www.cpaclicks.com/redirect.asp?a=xxxx&b=xxxxx&d=0&l=0&o=$landingpage-$uservote&p=0";
  $numchoices = 5;


?>

<html>

<frameset rows="1,*">

  <frame src="#">
  <frame src="<?php printf($promourls[rand(0,$numchoices)]); ?>">


</frameset>

</html>
Firstly you read in how they voted (yes or no) into the uservote variable. Then you set the landing page variable to whatever your landing page is (in this case vickpoll). Next you load an array with various different offers and insert the landing page and vote info into the subid field (this info can be accessed in COPEAC by downloading the optional info speadsheet in the stats area). Then select one of the offers at random from the array and load it into your frame. You can do a simple redirect to the url if you want but I prefer to keep my domain in the browser url field so I use frames (personal choice).

Simple as that. Review your conversion data and make changes to your offers based on your findings.
 
Also, is anyone using the content network or just search? I'm just getting a ton of empty impressions with no clicks from content, which is driving my CTR waaaaaay down. Does that have a big impact on my QS? Should I shut off the centent network?
Click to expand...
Content network has absolutely no bearing on your CTR (for search) or your QS.
You can see your CTR for search only in your account.

Dave or anyone, you getting any love from content at all?
 
Aequitas said:
Well to add to this big ass thread now, SEODAVE good idea and congrats on the profits, but one thing I would try out on your landing page is split testing your offers and monitoring clicks, it can be done easily with a database and a few lines of code, here I'll explain myself a little better.

You've got two buttons on your voting system, one is for YES and the other is for NO, instead of having them both point to the same offer have YES point to Offer A and have NO Point to offer B, now clock the links through Javascript or PHP (Depending on your hosting provider), now the user can't see the links in there status bar.

Ok if your doing a PHP re-direct type of clock, just before you redirect them increment a single value called clicks in a database table, then shoot the user out to the offer, it won't take longer then a few milliseconds to do.

Now you know how many clicks the YES button gets and you know how many clicks the NO button gets, and you've got web stats set up so you know how many impressions your getting, you also know how long people are staying on your site (Visit Length is good for re-working your designs to maximize stay times).

Anyway now you can easily split test two offers from the same landing page and you can figure out which converts more over a time frame, just set yourself a set amount of CTR before changing the offer, keep the better performing offer all the time and boom a bit more cash.

Everyone can even take this one step further with a radio button list poll, have the user select an option (1 out of 4 say), each different option brings them to a different offer, when they click submit, track the click (The radio button selected). Now you can track 4 offers at once instead of 2.

Tons of good info in this thread keep it coming. :)
Click to expand...

Aequitas - great minds think alike or something like that. I put this in place on Tuesday - we'll see how it converts - it does yes/no and the radio buttons too.

I like the iframe idea - perhaps mrkrongel can comment on putting aff offers in iframes . . .zip submits.

We'll see how it does, maybe I'll go krazyjosh and throw it up for sale, although I doubt anyone would pay for such a simple script.
 
I just duplicated this with a "infamous" guy, blammo, adwords is saying $5 minimum bids, fishing hell!@ Dead in the water, I will try yahoo and msn i suppose.

BTW, I have the guys name in the title, metas, info about him in another page on the site. Could it be because my landing page isnt linked from anywhere?
 
gnomic said:
OK. I just got PavarottiFuneral(dot)com. Given that I'd like to make money from the site and it's a rather sensitive subject to any potential visitors, I'm faced with a dilemma: put an AdSense banner at the top *or* along the sidebar... with my luck he'll be cremated and buried privately :(
Click to expand...

No line in the sand for you eh.
 
bit said:
No line in the sand for you eh.
Click to expand...

we're talking about making $$. as long as you aren't doing something that goes against your personal code of conduct / ethics, i don't see anything wrong with it. hell, if you don't see anything wrong with selling drugs, go for it, just know that you are taking a chance when you break the law. :yin-yang:

the person with the pavarotti funeral site obviously doesn't feel that is outside the realm of reasonable, with respect to his code of ethics, so fuck it. we're in this to make money, not please others! :rasta:

i think sean t. put it best:

i'm off the rack like slabs of ribs,
i want it big.
i ain't fucking with kids,
i'm after six digit figs.

:D:D:D

back on topic, i like emp's idea for testing as well. what i was trying to get across in my earlier post is that you need to have a control, something to gauge the rest of your results against.

personally, i think the best way to test this idea would be to send both the yes and no to the same offer. initiate a counter so each time the button is clicked, it sends the user to a different offer. try out maybe 5 different offers, the same for each button. this way, you will be able to see which offer converts best for which button. this is a conclusive result, which you can look at and decide the best course of action (obv in this case you would pick the best converting offer for each).

earlier this year, i was more involved with ppc than i am now. i was testing different landing page and site design, but i was changing too many variables at once. i was not able to draw any meaningful conclusions from these tests. sure, i could tell that by a) using more bulleted lists b) using larger fonts for headers and c) using sans seriff font converted the user better, but i couldn't point to exactly what was causing this reaction so i could duplicate in the future. by separating out these tests, and maybe, let's say, testing a few different styles of bulleted lists against each other, but only changing this one variable, i was able to determine which of these factors has a more of an impact on conversions and focus on optimizing this piece more (wow run on sentence). then i can move on to the next variable.

also recognize that different variables will interact and react with each other in ways that might not be readily apparent. for example, maybe larger headers with longer bulleted lists converts better than smaller headers with the same length of bulleted lists, but overall the smaller header converts better. :Yahoo_29:

these are just some things to consider when you are split testing. some of the pitfalls that i stumbled into would seem normal for beginners to get hung up on. hopefully you can gleam some knowledge on effective split testing from my post (if it isn't too incoherent).

i would love to hear some more experienced IM'ers chime in here on how they might test differently then i do, and if they worry more or less about the way that different variables will affect each other, depending on what variable you are currently testing.

hth
 
Status
Not open for further replies.
Top Bottom