Like everyone else looking for the solution to this problem, I googled it. I eventually found my way to a popular software program that you’ve probably heard of already called Xtreme Conversions. This was the first method I tried to successfully monitor my conversions.

This software basically generates a piece of code that you paste into your affiliate landing page which then records all the keywords users searched for. The trick with this software though is that it automatically appends either a TID (for clickbank products) or SID (for CJ products) to the click through. How this works is it gets the keyword from the AdWords ad, passes it through its software, records the keyword and assigns it a numerical ID. This numerical ID becomes your TIDs and your SIDs. When you make a sale via CJ, the SID numerical value will be assigned to the sale and will show up in your CJ sales reports. To find out which keyword generated that SID, log into your Xtreme Conversions page and it will show you a list of every keyword and its assigned ID. From this table, you can see that SID 45 for example, came from the AdWords keyword ‘Blue Widget’.

Each time a sale is made, an SID is appended to the sale and you can reverse track that to the keyword level to see which keywords are performing for you. Obviously you can cut the ones that arent.

Working with WordPress

For my affiliate sites I like to use WordPress. I like to use wordpress because its easy to update, has thousands of available free themes online and has a virtually unlimited stream of plugins and extras to help give my users a more comprehensive experience. Anyway, I found that if I used Xtreme Conversions (the none Pro version), that although it was tracking my keywords well, I couldn’t quite figure out how to set it up so that the code that it generates only needed to be pasted into my header.php file. If I only had one static landing page then the XC code and setup would have been perfect but because WordPress uses the same header include to generate every page, I found that it was generating an extra ‘frequency’ each time a page was loaded, even by the same user, and this distorted my stats. If you go and have a look at the Xtreme Conversions page, you will see some screenshots that will make more sense of what I’m talking about.

Anyway, to cut a long story short, the only way I could get it working properly with my wordpress pages was to either upgrade to the Pro version or install a php plugin to wordpress that would allow me to insert php code directly to posts. I tried the php plugin but it still did not work as intended and by that point, I was fed up with faffing around so looked for another alternative.

Simple, Free, Effective Way to Track AdWords Keywords via CJ

I went back to google and tried again. This time I found an old thread on the Digital Point forum that showed me what I needed to do in a nutshell. Now although this didn’t work for me right out of the hat, it did lead me to the location where I was finally able to put the whole thing together and this is how I did it:

1. Log into your AdWords campaign and setup all of your different ads so that they pass the exact keyword search typed in by the user to get to you. With Google AdWords, you can use the Dynamic Keyword Insertion to achieve this goal like so:

Your adwords destination url:

http://www.yourdomainhere.com/yourlandingpagehere.php?kw={keyword}

You can see the bold at the end. This is what you need to append to the end of your destination url in your adwords text ad/ad. What this does is it will add the {keyword} the user searched for to the end of the string so that the url passed to your page will be:

http://www.yourdomainhere.com/yourlandingpagehere.php?kw=your keyword

Now what we need to do is set up a way to pass that keyword to CJ by adding an SID to the end of all of our outgoing CJ affiliate links. So that the outgoing link would look like this:

http://www.cjlinkwouldgohere.com/cjlandingpageorproducthere.php?SID=keyword

This can be achieved with the following php script:

<?php
$keyword = $_GET[’q'];
if(!strcmp($keyword, “”))
{
echo ‘<a href=”http://www.anrdoezrs.net/click-2412142-10482525?sid=Default”>Anchor</a>’;
} else {
echo ‘<a href=”http://www.anrdoezrs.net/click-2412142-10482525?sid=’ . $keyword . ‘>Anchor</a>’;
}
?>

You should re-write this code because if you copy and paste some characters such as the ‘ or ” can get mixed up by your browser/my page. If all of your links are being generated in this way, then you will be able to track your keywords via CJ and see which ones are making you money.

If you are using wordpress like I do, then the plugin you need to be able to insert this code where your link needs to go is called phpexec. Just write the code inside a blog post or page but make sure the code is inside some <phpcode> </phpcode> tags.

My thanks go to Jim Karter who helped me track my adwords conversions with CJ on his original post here.

Hopefully it will help you too!

Yesterday when I started the campaign, I had a Quality Score of Great 9/10 for both of my Ad Groups on 95% of their keywords. If you want to know how to improve on this, see my post on how to get an AdWords Quality Score of 10/10. Anyway, back to the point at hand, the campaign had been live for about 10 hours and was achieving a CTR of about 4.5% but my conversions on my landing page didn’t seem to match. As always, I felt the urge to try and tweak things a little but am now considering the possibility that the tweaks I made to my landing page have affected the Quality Score of the campaign adversely. So what did I change?

Well, first up I was trying to write a good, rich content landing page, with plenty of correct SEO techniques. The page was well written, 100% unique and made appropriate use of H1, H2, H3 tags, emphasis, underlines and bold. It was written for the human, not the search engine and it was completely on topic, relevant and semantic.

Now, when I first launched the AdWords campaign, I had a quality score of 9/10 and a minimum bid 0.02 to be on the first page. This seemed to stay for around 8 hours, during which time I made an adjustment to my H1 tag and added 3 prices to 3 product reviews. So, for instance, let’s assume my original header read:

You’re here because your brand new HD TV with built in Blu-Ray just lost its picture on Sky

And the new header:

Learn how to fix your Blu-Ray HD TV picture in just 60 minutes from your own home

Now this is not the niche I am in, and I have tried to emulate the concept of the change as accurately as possible. If you can imagine that prior to the change, my keywords (all based around ‘fixing HD TV’ terms) had a Quality Score of 9/10 and then after the change (although I can’t be sure it was the change in H1 tag that caused it – this is why I am investigating it here!) and about 8 hours, it had dropped to only 7/10.

Could the small H1 tag have such a dramatic effect on Quality Score?

I am going to investigate this in three ways:

1. I am going to test the theory on an existing AdWords keyword that has had a solid QS of ‘Great’ for a couple of months and a Quality Score of 9/10 and just slightly adjust the landing page H1 text to still be relevant but in a different way – Smells like LSI ?
2. I am going to create a new landing page using the same philosophy as the one that caused this whole investigation and not change anything for over a week.
3. I am going to create a second landing page with a not so good (in my eyes) quality score to see if new campaigns get some sort of quality score boost until google has a good chance to investigate them further.

I did mention another change earlier, but I fail to see how this could affect the quality score at all. Basically, there was 3 product reviews on the site and initially, they had a price just above the affiliate link to ‘prepare’ the user for the price when he lands on the merhcant’s product page I was promoting. I removed these prices completely to let the affiliate find the price after he’s clicked on my link (I’ll leave the conversion up to the merchant, it’s their job to create high converting pages – at least that was my thoughts).

Other than these two changes (the H1 tag and the price removals), no other changes were made to the landing page so either the quality score dropped because of 1 of these changes or it dropped due to time. I will post again once my investigations are complete.

Afterthought: Once I saw the QS had dropped from 9/10 to 7/10, I also added links to a ‘Privacy Policy’ and ‘About This’ page, just in case it was a time-related factor and once google figured out I was promoting a product as an affiliate, they tightened up their hold on me.