Double Digit CTR, wtf!?

Today I decided to try something new with my Google Adwords PPC ads.

I added 500 keywords to the SAME adgroup and gave them individual bids. That’s pretty crazy for me considering I usually have ONE keyword (with 2-3 matching types) per adgroup. And so far it’s been working brilliantly for me.

But today, I decided to give the multiple-keywords-in-an-adgroup trick a go–it’s part of my original 1,000 leads/day challenge I set maybe a month or so ago (if you remember).

Today I decided to try something new with my Google Adwords PPC ads.

Hear this story first…

A few weeks ago I was guest of an online Radio Show courtesy of my pal Ben Settle.  The main topic was Google Adwords and I was brought on to talk about my experience with PPC. While on the call, the other guest showed a snapshot of his adwords account where he was bidding on over 400 keywords–get this–in the SAME adgroup. That’s something a lot of experts, including Perry Marshall will tell you is a biiiiig no-no.

Well the surprising part that had me baffled was that he had a 93% conversion rate. Dude, there’s the very same screenshot I saw in the captioned image below. What a guy! Seriously, I have been impressed since and that’s a really long time to be impressed with someone, ok?

Today, I thought finally to give it a go myself.

I added 500 keywords to the SAME adgroup and gave them individual bids. That’s pretty crazy for me considering I usually have ONE keyword (with 2-3 matching types) per adgroup. And so far it’s been working brilliantly.

But I decided to give the multiple-keywords-in-an-adgroup trick a go. After all, I’m still working on my original 1,000 leads/day challenge that I set maybe a month or so ago (if you remember).

This Adwords Account Snapshot Had Me Baffled: 400+ Keywords In The Same AdGroup With a 93% Conversion Rate With A Landing Page That Should Never Have Converted!
This Adwords Account Snapshot Had Me Baffled: 400+ Keywords In The Same AdGroup With a 93% Conversion Rate With A Landing Page That Should Never Have Converted!

One of the lessons I learned through my online marketing experience is that just because something is working doesn’t mean it’s the best way. So what were the results of this experiment?

First, my click-thru-rate on that adgroup went from 1.5% to 12%. Heck yes! Double digit CTR!

That doesn’t even make sense. Considering that i’m using the exact same ad on 500 keywords. Theoretically, the more targeted the ad is to the keyword, the better response it should get. Here, instead, 500 different keywords with the same (much less tailored ad) are getting clicked more frequantly. Here’s why I think this happened:

When bidding on just a single keyword I would use the same bid for potentially 1000’s of searches. But when I’m targeting longer-tail keyword derivatives I’m gaining the advantage of being able to put down individual bids more customised to each search string’s competition. And I also think that Google is giving preference to my ad over others, regardless of my bids, because I’m telling it specifically (instead of implying) that it should show the ad on those exact search strings.

There is a down-side to this trick though.

My conversion rate shot all the way down. I mean, there’s literally ZERO conversions for the couple of hundreds clicks accumulated today.

Expensive experiment. But a very valuable lesson learned about increasing click-thru-rate.

Now, onto figuring out how to make the landing page convert more of those clicks!


To learn more of Jim’s PPC advertising techniques, check out PPC Domination.

Author: Jim Yaghi

Jim Yaghi is an advertising consultant and traffic expert, with a background in Artificial Intelligence.

38 thoughts on “Double Digit CTR, wtf!?”

  1. @Rebecca Geiger haha! if you could do the hose trick, that would be easy 😉

    @Forrest Bivens sure man..basically when you bid on a keyword, eg “mlm” in phrase match for example, you’re matching mlm leads, mlm home business, mlm training, mlm companies, best mlm, etc. That’s TONS of other searches that will be matched but you’re short-cutting with “mlm”. You’re implying that it should match those searches.

    But if you actually specified and wrote out individuall all of those keywords as well as “mlm” phrase match (bid a little lower on mlm so that the other keywords are given a chance), then you’re telling google “Hey i want you to show my keywords on these explicit keywords!” So it prefers you over other people who didnt specify them and just used “mlm”.

  2. Hi Jim,

    Would you mind clarifying this part?

    And I also think that Google is giving preference to my ad over others, regardless of my bids, because I’m telling it specifically (instead of implying) that it should show the ad on those exact search strings.

    Thanks man,

    Forrest

  3. Hi Jim,

    Would you mind clarifying this part?

    And I also think that Google is giving preference to my ad over others, regardless of my bids, because I’m telling it specifically (instead of implying) that it should show the ad on those exact search strings.

    Thanks man,

    Forrest

  4. Ahhhh, got it! Thanks!

    So, in this example, if I used all the longer tail keywords only & not use "mlm" at all, do you think my ad would come up on a search for mlm? Because it seems that it works this way when you are using "mlm" and someone searches for say mlm training, mlm companies, best mlm, etc…assuming google gives you a good quality score.

  5. Ahhhh, got it! Thanks!

    So, in this example, if I used all the longer tail keywords only & not use “mlm” at all, do you think my ad would come up on a search for mlm? Because it seems that it works this way when you are using “mlm” and someone searches for say mlm training, mlm companies, best mlm, etc…assuming google gives you a good quality score.

  6. Nah, you're gonna need to bid on [mlm] at least. The others match because "mlm" says match strings that include mlm. but "mlm leads" says match searches that contain mlm leads such as mlm leads, mlm leads training, live mlm leads, etc.

  7. Nah, you're gonna need to bid on [mlm] at least. The others match because “mlm” says match strings that include mlm. but “mlm leads” says match searches that contain mlm leads such as mlm leads, mlm leads training, live mlm leads, etc.

  8. Thanks again Jim…actually I mis-spoke. I should have asked the same question using a broad phrase (mlm without quotes).

    You da man… :)

  9. Thanks again Jim…actually I mis-spoke. I should have asked the same question using a broad phrase (mlm without quotes).

    You da man… :)

  10. Hi Jim,

    I know that CTR is important, but aren't conversions of primary importance? To me, you are treading into territories that most of us can't afford to, so I appreciate you sharing your experiences with the rest of us.

    I will be watching closely to see if this is the kind of information we can all use to our benefit. Right now, the experiment seems a little out of my league, but that may change soon.

    Lane Reiss

    1. Of course conversions are of primary importance. Can you imagine, though, would you get any conversions without clicks? I run the three elements of successful PPC experiment in three ISOLATED and sequential phases: 1) Impressions, 2) Clicks, 3) Conversions.

      So if you got a high CTR you got 2/3rds of your experiment done. Unless you've made a ridiculously dishonest promise in your ad, the last phase should be a matter of playing with your landing page and how you word it.

      Everyone, no matter how limited their budget could (and should) run such experiments. This was only "expensive" because I left it running and didn't care that it didn't convert. I wanted to conclusively be able to say that my landing page was flopping (it was a new one never tested before). When you consider it cost about $30 to find out forever how to get double-digit-ctr, i think it would have been worth-while for any of you.

      Jim

  11. Hi Jim,

    I know that CTR is important, but aren't conversions of primary importance? To me, you are treading into territories that most of us can't afford to, so I appreciate you sharing your experiences with the rest of us.

    I will be watching closely to see if this is the kind of information we can all use to our benefit. Right now, the experiment seems a little out of my league, but that may change soon.

    Lane Reiss

    1. Of course conversions are of primary importance. Can you imagine, though, would you get any conversions without clicks? I run the three elements of successful PPC experiment in three ISOLATED and sequential phases: 1) Impressions, 2) Clicks, 3) Conversions.

      So if you got a high CTR you got 2/3rds of your experiment done. Unless you've made a ridiculously dishonest promise in your ad, the last phase should be a matter of playing with your landing page and how you word it.

      Everyone, no matter how limited their budget could (and should) run such experiments. This was only “expensive” because I left it running and didn't care that it didn't convert. I wanted to conclusively be able to say that my landing page was flopping (it was a new one never tested before). When you consider it cost about $30 to find out forever how to get double-digit-ctr, i think it would have been worth-while for any of you.

      Jim

  12. Hey Jim,

    I wanted to ask you when you run a campaign how many clicks does it take a keyword in order to find out if it converts or you decide to just drop it. I just wanted to know how long do you wait to make a decision.

    1. Keith,

      It's gonna really depend on a lot of factors…particularly what you set aside as a testing budget. I personally like to make sure my experiments are conclusive, and the more data about them the better.

      Typically, i like to run an experiment unchanged for a full 24-hrs. Sometimes, the experiment is wayyy too expensive to run for a not-so-worthwhile result. So i won't let it run its course.

      If a split-test IS in fact converting, i like to run it a full week. As more data comes in, you will find the "winner" changes and stabilises with more data.

      As a very very loose rule of thumb, if the keyword you're bidding on costs you double what it's worth to you in revenue, kill the keyword and move on to another keyword.

      Jim

  13. Hey Jim,
    I wanted to ask you when you run a campaign how many clicks does it take a keyword in order to find out if it converts or you decide to just drop it. I just wanted to know how long do you wait to make a decision.

    1. Keith,

      It's gonna really depend on a lot of factors…particularly what you set aside as a testing budget. I personally like to make sure my experiments are conclusive, and the more data about them the better.

      Typically, i like to run an experiment unchanged for a full 24-hrs. Sometimes, the experiment is wayyy too expensive to run for a not-so-worthwhile result. So i won't let it run its course.

      If a split-test IS in fact converting, i like to run it a full week. As more data comes in, you will find the “winner” changes and stabilises with more data.

      As a very very loose rule of thumb, if the keyword you're bidding on costs you double what it's worth to you in revenue, kill the keyword and move on to another keyword.

      Jim

  14. I was reading ore of Jims site when this question came in, so I hope I'm not crossing the line when commenting before Jim (THE Jedi) does?

    I (personally) take into consideration the Impression, CTR & CPC. Assuming you did some decent keyword research beforehand & validated it as a good keyword, clicks & CTR in general will then depend on your adcopy. Conversions will depend on your landing page quality.

    I await the Jedi's advice to see if I am off in the outback or closer to town on this one! :)

    1. Forrest,

      Yeah man, that's right. there's no real quick-and-easy way to tell if it's time to kill a keyword or experiment. The ultimate test is the answer to this question:

      "Is making this keyword work worth the money and time i'm spending on making it work?"

      A keyword with high volume of search and good quality of searcher may well justify spending hundreds if not thousands in testing budget.

      But if it's a keyword that's gonna ultimately bring you few qualified searches, and you're going to make a total of $10 on the eventual sale…leave it alone and move on. There's an infinite supply of keywords and you're bound to find something better and easier.

      jim

    2. Keith,

      the length of time you have an ad up costs nothing if it's getting no clicks. So the test isn't time…nor is it clicks.

      It's COST.

      Suppose you're earning $20 in commission on a sale conversion, then you should never test a keyword longer than the total spend of $40 on testing it. If it's proving THAT difficult to convert, there's no point wasting your resources on it. Move on, there's definitely another keyword easier to convert out there.

      Jim

  15. I was reading ore of Jims site when this question came in, so I hope I'm not crossing the line when commenting before Jim (THE Jedi) does?

    I (personally) take into consideration the Impression, CTR & CPC. Assuming you did some decent keyword research beforehand & validated it as a good keyword, clicks & CTR in general will then depend on your adcopy. Conversions will depend on your landing page quality.

    I await the Jedi's advice to see if I am off in the outback or closer to town on this one! :)

    1. Forrest,

      Yeah man, that's right. there's no real quick-and-easy way to tell if it's time to kill a keyword or experiment. The ultimate test is the answer to this question:

      “Is making this keyword work worth the money and time i'm spending on making it work?”

      A keyword with high volume of search and good quality of searcher may well justify spending hundreds if not thousands in testing budget.

      But if it's a keyword that's gonna ultimately bring you few qualified searches, and you're going to make a total of $10 on the eventual sale…leave it alone and move on. There's an infinite supply of keywords and you're bound to find something better and easier.

      jim

    2. Keith,

      the length of time you have an ad up costs nothing if it's getting no clicks. So the test isn't time…nor is it clicks.

      It's COST.

      Suppose you're earning $20 in commission on a sale conversion, then you should never test a keyword longer than the total spend of $40 on testing it. If it's proving THAT difficult to convert, there's no point wasting your resources on it. Move on, there's definitely another keyword easier to convert out there.

      Jim

  16. hmmm,

    So Jim, its best to let it run BUT if its costing me too much money and I see no results I should drop it? For someone that doesnt have THAT big of budget how long should I test a keyword? Should I measure by clicks or just time?

  17. hmmm,
    So Jim, its best to let it run BUT if its costing me too much money and I see no results I should drop it? For someone that doesnt have THAT big of budget how long should I test a keyword? Should I measure by clicks or just time?

  18. Jim,

    For one of my biz-ops, the commissions are so high on a sales conversion I could run a keyword till hell freezes over before giving up on it. So the decision to pull the plug or keep a keyword is all based on impressions, ctr, cpc (this conversion is defined by them becoming a lead/no lead.) The post lead conversion process with this particular biz-op really determines the sales conversion…

    So your self-questioning regarding 'is it worth the effort' REALLY takes precedence. Heck after a while, it has become clear what keywords work & which are duds.

    Riding those thoroughbreds till they drop & have killed the nags. Split-testing & tweaking all the while to squeeze all I can outta the thoroughbred keywords!

    Man I like your mind my friend… signed,

    Jedi-in-the-making 😉

  19. Jim,

    For one of my biz-ops, the commissions are so high on a sales conversion I could run a keyword till hell freezes over before giving up on it. So the decision to pull the plug or keep a keyword is all based on impressions, ctr, cpc (this conversion is defined by them becoming a lead/no lead.) The post lead conversion process with this particular biz-op really determines the sales conversion…

    So your self-questioning regarding 'is it worth the effort' REALLY takes precedence. Heck after a while, it has become clear what keywords work & which are duds.

    Riding those thoroughbreds till they drop & have killed the nags. Split-testing & tweaking all the while to squeeze all I can outta the thoroughbred keywords!

    Man I like your mind my friend… signed,

    Jedi-in-the-making 😉

  20. Jim,

    After looking at the snapshot you gave with this, I was wondering if his CTR had to do with him using Dynamic Keyword Insertion in both his headline and Body of the Ad (line #2).

    There are definitely times when I group a tight "theme" of keywords (long tail variations) together in the same adgroup, and use DKI for the ads. I've done this in the past just to see if a keyword has potential and gets clicks. I've done this alot with long-tail keyword combinations.

    Using the example you cited above, grouping together variations of "mlm leads" its possible to group several of them in the same adgroup and use DKI.

    What is your experience with dynamic keyword insertion? Do you recommend trying it or staying away from it?

  21. Jim,

    After looking at the snapshot you gave with this, I was wondering if his CTR had to do with him using Dynamic Keyword Insertion in both his headline and Body of the Ad (line #2).

    There are definitely times when I group a tight “theme” of keywords (long tail variations) together in the same adgroup, and use DKI for the ads. I've done this in the past just to see if a keyword has potential and gets clicks. I've done this alot with long-tail keyword combinations.

    Using the example you cited above, grouping together variations of “mlm leads” its possible to group several of them in the same adgroup and use DKI.

    What is your experience with dynamic keyword insertion? Do you recommend trying it or staying away from it?

  22. Kurt,

    definitely that result can't be reproduced just by using Dynamic Keyword Insertion. I have used it before but i don't like that it takes control away from me. For my own ads, this would be just a test phase. I would write an ad that uses the seed keyword and that's all.

    You can see in follow-up experiments, I got rid of most of the keywords.

    Over time, I'll be phasing them back in, but each with its own adgroup and ad combination.

    If you like DKI, then go ahead and use it. I just happen to be one of the people that doesn't like it.

    Jim

  23. Kurt,

    definitely that result can't be reproduced just by using Dynamic Keyword Insertion. I have used it before but i don't like that it takes control away from me. For my own ads, this would be just a test phase. I would write an ad that uses the seed keyword and that's all.

    You can see in follow-up experiments, I got rid of most of the keywords.

    Over time, I'll be phasing them back in, but each with its own adgroup and ad combination.

    If you like DKI, then go ahead and use it. I just happen to be one of the people that doesn't like it.

    Jim

  24. Hi Jim,

    Thanks for all your time. I've just been going through your PPC course. I was wondering how do you know if a certain keyword converts? Say for example if someone clicks on your ad from one of your google adwords groups (I have a few groups all with the same URL). I see that I have a 1% CTR from one adgroup and a 2% CTR from another and both have good performing keywords and both ads, go to the same site. How do I know which ad converted? i can tell if they clicked on my landing page but not which ad they came from to make them convert. Do we make a statistical guess because of ther CTR?

    Thanks

    kathryn

  25. Hi Jim,

    Thanks for all your time. I've just been going through your PPC course. I was wondering how do you know if a certain keyword converts? Say for example if someone clicks on your ad from one of your google adwords groups (I have a few groups all with the same URL). I see that I have a 1% CTR from one adgroup and a 2% CTR from another and both have good performing keywords and both ads, go to the same site. How do I know which ad converted? i can tell if they clicked on my landing page but not which ad they came from to make them convert. Do we make a statistical guess because of ther CTR?
    Thanks
    kathryn

Leave a Reply

Your email address will not be published. Required fields are marked *