The importance of tracking when paying for clicks really can’t be overstated. Without a very clear idea of which keywords are converting into cash, you’re really flying blind. By my own admission, I’ve been pretty lax in regards to keyword tracking until only recently.
I realize there are some handy scripts around that make this a fairly easy task, but if you can live without the built-in reports, keyword tracking is really quite easy to implement yourself with just an Excel spreadsheet and a small bit of PHP code.
In a future post I’ll describe how to do that for anyone unfamiliar with the procedure, but for now, I’d like to point out another conversion metric associated with keywords that is often left unmonitored — clickthrus to the promoted offer.
In my case, the offers being promoted are reached via affiliate links.
Over the past several months, I’ve been tracking what’s known as “out clicks” from my landing page. This is simply watching what users do after arriving on my landing pages.
Do they bounce, i.e., hit the back button immediately after arriving?
Do they hang around for a bit and then navigate away without clicking on an affiliate link, dammit?
Or do they at the very least do me the courtesy of clicking thru to the affiliate offer, thank you very much?
At the end of the day, unclicked affiliate links don’t have a snowball’s chance of earning me a commission.
Yep, pretty obvious, but do you know, which of your PPC keywords aren’t giving you a fighting chance? The keywords that are just burning up your ad spend and for whatever reason aren’t bringing visitors receptive to what you have to offer on your landing page.
Analytics software such as Google Analytics, as comprehensive and invaluable as it may be, does not show you out clicks. Your server logs are also no help in this regard. Thus far the only analytics software that I’ve managed to find that records and displays out clicks is Site Meter. I have tried StatCounter, but I find it’s interface a bit clunky and couldn’t find if it displayed out clicks. Someone please correct me if I wrong about this.
Site Meter, like a lot of applications and services these days, comes in both a free and paid version.
So far I’ve only been using the free version and the information I get from it is more than enough to keep me busy. The free version only displays the last 100 visits, but as often as I check it, that’s plenty of history.
The free version also forces you to display an image on your landing page, but I just choose the tiny square without any text and locate it discreetly in either the page footer or at the bottom of the sidebar.
Also be sure to set your privacy level to the highest so nosey visitors can’t view all your analytics data. For some reason the default privacy level is set to the lowest level.
I check my Site Meter stats whenever I need a mental break from doing other stuff.
I check it more frequently immediately after launching a PPC campaign. I can see visits well before they show up in my Adwords admin panel. So right from the get-go I can see if I’m getting impressions for my ad, which keywords are getting clicks, where my ad is ranking, and most importantly, is are my affiliate links getting clicks?
I set up multiple Site Meter accounts for my sites and campaigns in order to keep things from getting too cluttered. Yes, this is a bit of a pain, but if you use FireFox, then the auto-complete and password remember features make logging in to multiple Site Meter accounts a whole lot easier.
This is the intermediate stage of campaign tracking prior to any conversion action that may take place on the merchant’s site. With this data, I can immediately start collecting negative keywords to roll back into my Adwords campaigns. Conversely, I can also get ideas for longer tail keywords to add in based on the broad and phrase match keyword types that triggered the ad impression.
For a campaign that I’m currently running, I discovered firsthand the insidious side effects of Google’s “synonomous matching“. This is not to be confused with their even more insidious “related matching“.
Synonymous matching occurs when you’re using broad match types with your keywords. In general, I subscribe to the “go wide” before “going deep” approach to keywords and ad groups promoted by Amit Mehta, who is certainly no slouch when it comes to Adwords. That’s why it’s recommended that you use all 3 match types initially.
However, if you explicitly want just “smooth green widgets” because you only want to target buyers who are interested in “smooth green widgets” and you don’t want to waste clicks on the unwashed masses looking for just “green widgets“, well guess what?
Google tries to get a little too helpful and shows your ad for “green widgets” searches even though you don’t have this search term for any match type in your ad group. But since you have “smooth green widgets” with the broad match type, they just go head and display your ad anyway.
I found this out from monitoring my landing pages with Site Meter and as to be expected, the visitors who were searching for only “green widgets” hit my landing page and probably went “eew, smooth green widgets – I’m outa here!“.
Then why the heck did you click on my ad when it clearly said “smooth green widgets” you moron? Some people!
So being able to spot these trends very early in a PPC campaign can give you a head start and hopefully give your ROI a needed boost.
on Aug 22nd, 2008 at 12:00 am
[...] | user-saved public links | iLinkShare 2 votesAdwords Analytics With Site Meter>> saved by jaquinta 1 days [...]