• http://www.facebook.com/profile.php?id=738199271 Joseph Sanchez

    Great article. So much to learn with PPC! Thanks for the article Carrie

  • http://www.andykuiper.com Andy Kuiper – SEO Analyst

    Interesting, thanks Carrie. I appreciated the step-by-step instructions, they easily broke down a complicated set of manoeuvres into a number of easy steps ;-)

  • http://twitter.com/GnosisArts Gnosis Media Group

    Are you sure? Because I’ve read in more than one place that ad position really has no statistically significant effect on ROI for the vast majority of advertisers. But I’m intrigued now, so I want to look deeper into this one to form my own conclusion.

  • http://twitter.com/GnosisArts Gnosis Media Group

    If the premise of this post is correct, then the old theory that ad position is not significantly related to conversion rate (http://adwords.blogspot.com/2009/08/conversion-rates-dont-vary-much-with-ad.html) or revenue may be dunked?

  • http://www.facebook.com/carrie.hill1 Carrie Hill

    I think ROI is greatly affected by position – REVENUE is a different number than return. if you spend less for lower positions, but make the same or very close to the same money – your return on investment is much higher than paying a lot for top positions.

    The numbers show that revenue based on position can vary for some keyword phrases…that’s shown clearly in the screenshots above.

  • http://twitter.com/GnosisArts Gnosis Media Group

    There’s a lot of meat in here, Carrie. I’ve been studying this short post for that past hour, and it’s starting to make my head spin a bit. Question: Would Either Enhanced CPC or Conversion Optimizer probably do this also?

  • http://twitter.com/GnosisArts Gnosis Media Group

    Awesome stuff. I see what you mean. I’m beginning to think that the blanket statement, “Conversion Rates Don’t Vary Much with Ad Position” (which is what the Adwords people would have us believe) is no longer valid – if it ever was. Because, as I’m looking at my keywords now, I see that conversion rate is definitely varying, quite dramatically in some cases, by ad position.

  • http://twitter.com/meseostrategist Rahul SIngh

    helpful article for new learner as well in PPC and Analytic integration 

  • http://twitter.com/xnumerik xnumerik

    My knowledge of AdWords and Analytics is still basic but reading such articles motivates me to learn a lot more to make the most out of them, thx for the great tips, I have tweeted this link too!

  • Pat Grady

    We manage client PPC on an ROI-centric basis, but every once in a while, a client asks us “why aren’t we in first position for Red Widgets?”…  thinking of changing my explanation / discussion to “sure, which better converting keyword would you like me to sacrifice to make this happen?”  Bahaha!

  • PeckLorrie254

    Online Money Making is one of the most demanded things from couple of years. Everyone wants to earn money efficiently and with little efforts. There are so many ways by which you can earn money online, just some techniques and little effort is required to do. Right now internet is one of the richest fields where one can not only earn money also can prove him by making his dreams true. One of my best friend is earning good weekly cash and last week he got 2000$. Just move here for further details http://earnusdathome.weebly.com/

  • http://www.rimmkaufman.com/ George Michie

     This is not an “old theory”, it is an old fact based on careful study of tons of data, and done with an understanding of statistical variance.  This post totally misses the boat.  Click volume varies materially with position.  Revenue per click does not.  Cost per click varies terrifically by “position.”  The proper way to bid for maximum revenue within an efficiency constraint is to measure the revenue per click and bid to a comfortable fraction of that number to hit efficiency targets.  Position is as much determined by competitor’s bids as your own.  Position bidding was a bad idea before personalization made the concept of position irrelevant.  Now it’s just absurd.

  • http://twitter.com/jasmineJaz2 jasmine (Jaz)

    good information about  Google analytic,ppc thanks for sharing

  • http://www.makeitbloom.com/xurxo-vidal/ Xurxo Vidal

    I agree with George Michie that ad positions impact traffic volume rather than revenue per click. Revenue per click is directly influenced by keyword selection and ad copy in combination with landing page content and experience.

    Lowering positions can yield a better ROI because conversion rates will dictate revenue per click. The higher the conversion rate, the more you can afford to bid to generate traffic profitably.  It would have been interesting to see if there were any significant fluctuations in conversion rates based on ad positions (provided that a particular keyword got a similar amount of exposure in each position).

    Also there’s the question of statistical significance. Any optimizations based on too little data might likely backfire or cause missed revenue. Could it be that certain keywords generated higher revenues in lower positions in the examples shown because they more often triggered ads to appear in lower positions based on what the bids allowed (and more rarely showed up in higher spots)?

  • http://twitter.com/GnosisArts Gnosis Media Group

    We have implemented Carrie’s suggestions above this week and we’ve already seen increased conversions. Too early to tell if there is significance, obviously, but it certainly looks promising.

    Second, you said ”  Revenue per click does not [vary by ad position].” But how can this be? If Ad A in Position 1 produces a conversion rate of 20%, but only produces a 2% conversion rate in position 10, then it doesn’t take an expert statistician to figure out that Position 1 is probably generating more revenue that Position 10.

    So, why is Position Bidding a bad idea again?

  • http://www.rimmkaufman.com/ George Michie

     Exactly right.  It is not easy to study precisely because if you bid smartly you’re likely higher on the page for KW that convert better, which can make folks believe RPC varies.  RKG, EF, Marin, The Search Agency and Google themselves have all reached the same conclusion.  We found a slight decline in conversion rates in positions 1 and 2 when they are above the organic listings, and a slight increase in conversion rates at the bottom of the page, but too slight to impact smart bidding systems (within the error bars of RPC calculations for almost any ad).  If you’re talking about a KW like “Car Insurance” where the advertiser might spend a couple million bucks a month on a single KW, it’s worth playing with, but personalization of results make the averages meaningless.

  • http://www.rimmkaufman.com/ George Michie

     For ecommerce companies that take sales on the website it’s simply revenue driven divided by the number of clicks.  Or conversion rate times average order size.  If you’re limited to Google’s conversion tracking conversion rate is the best proxy.  In all of paid search the reality is that there may be exceptions to the rule.  Best of luck to you.

  • http://twitter.com/jonintweet Joni S.

    Hey Carrie,

    this is an excellent approach to determining keyword profitability. The problem with ‘cost per conversion’ is indeed that it neglects the REVENUE of conversion.

    However, the explanation behind the strange behavior (revenue differs based on position) may be simply that those customers bought more. If this is the case, the return of a position is actually affected by a random variable, which is the different consumption behavior of customers clicking. In consequence, setting bids to average positions (like 8th) would actually not be relevant at all, as the data is not based on consistent but random behavior.

    What do you think?