Adwords Experts, When Do You Get Rid Of a Keyword?

7 replies
  • SEO
  • |
I am interested in learning which criteria you guys/girls use to determine when a keyword gets paused. For instance I have a keyword that has 36 clicks with a total cost of 12 bucks. I get $14 per sale and that keyword made me 0 sales so far. Should I let the keyword get at least 100 clicks to test it properly or should I look at amount of money spent. Let's say, if I spent $30 on it and no sales it needs to get paused?

Another Scenario I keep running into is having a keyword with 234 clicks at a cost of $110 and made sales for $130 leaving me with $20 in profit. I can't get cpc lower on it. Do I keep the keyword or let it go?

Any advice and tips would be much appreciated.

Cheers
#adwords #experts #keyword #rid
  • I´m not an Adwords expert, but here are my thoughts.

    In the first example you have to make at least 1 sale with about 42 clicks to get even. To make a profit you have to sell more. Obviously it´s even very difficult to get even and making a profit would be nearly impossible. So I would get rid of that keyword.

    The second one is making a profit. I would keep everything that´s making a profit. If you cannot get cheaper clicks, may be you can improve the conversion. If not, it´s still $20.
    {{ DiscussionBoard.errors[2195016].message }}
    • Profile picture of the author jan roos
      On the first scenario, what if the keyword makes 3 sales in between 36 clicks and 100 clicks? That would mean it's a profitable keyword. That's why I am asking if anybody has specific criteria for when they would pause a keyword as 30 clicks might not be enough to test properly.
      Signature

      I'll teach you how to make money like a Mamba.

      Sign up for the free money mambas newsletter!

      {{ DiscussionBoard.errors[2195972].message }}
      • Profile picture of the author digitalcostas
        Hey Jan this is a really great question... these is the sort of things I never see in 99% of the fancy IM/Adwords courses and probably the most important.

        Statistically speaking (I believe I heard Glenn Livingston state this) if you go 32 clicks with no conversion something is wrong.

        That being said, the tough thing about Adwords (and marketing in general) is the timing... keywords that don't convert today may be winners sooner or later.

        For immediate purposes, especially if you're blowing through your budget, I'd pause that word and invest that money trying to find other keywords to convert in < 32 clicks.

        If you're in the testing mood and not under pressure I might try some different things... remember each individual *keyword* represents a unique conversation that's going on in that searchers brain RIGHT NOW... that keyword might convert if the ad copy/landing page was different.

        I'd try putting up a two-question survey as the landing page for that keyword asking "What's your single biggest worry/question/frustration about topic?" and also "What happened that caused you to search for TOPIC in Google just now?" and give it 32 more clicks to see if you get any meaningful responses...

        -Costas
        {{ DiscussionBoard.errors[2196020].message }}
        • Profile picture of the author jan roos
          Originally Posted by digitalcostas View Post

          Hey Jan this is a really great question... this is the sort of things I never see in 99% of the fancy IM/Adwords courses and probably the most important.

          Statistically speaking (I believe I heard Glenn Livingston state this) if you go 32 clicks with no conversion something is wrong.

          That being said, the tough thing about Adwords (and marketing in general) is the timing... keywords that don't convert today may be winners sooner or later.

          For immediate purposes, especially if you're blowing through your budget, I'd pause that word and invest that money trying to find other keywords to convert in < 32 clicks.

          If you're in the testing mood and not under pressure I might try some different things... remember each individual *keyword* represents a unique conversation that's going on in that searchers brain RIGHT NOW... that keyword might convert if the ad copy/landing page was different.

          I'd try putting up a two-question survey as the landing page for that keyword asking "What's your single biggest worry/question/frustration about topic?" and also "What happened that caused you to search for TOPIC in Google just now?" and give it 32 more clicks to see if you get any meaningful responses...

          -Costas



          -
          I like what you've said except for that if I get 32 clicks and no sales something is wrong because I've had keywords that had 30 something clicks and no sales come back and make 3 quick sales which turned it profitable.

          I like to give my keywords at least 100 clicks to do proper testing but because I have a lot of keywords it can get quite expensive. Also there is the issue of some keywords just being barely profitable which makes me think I could have spend that money on articles or Senuke or something.

          Cheers
          Signature

          I'll teach you how to make money like a Mamba.

          Sign up for the free money mambas newsletter!

          {{ DiscussionBoard.errors[2196066].message }}
          • Profile picture of the author digitalcostas
            Originally Posted by jan roos View Post

            I like to give my keywords at least 100 clicks to do proper testing but because I have a lot of keywords it can get quite expensive.
            Yeah I hear you... i don't like it either - I wish situations weren't so mutually exclusive and we could deal with empirical facts instead of fuzzy rules of thumb.

            But... it also puts us in a better place to compete and win in the long run because most regular folks running adwords either a) don't think this deep about their campaign or b) have no idea how to even begin answering that question. At least we've got some theories.

            Also you might try some other factors too... have you tested different price-points? How long are the people who click and not buying staying on your landing page- long enough to read the sales copy?

            -Costas
            {{ DiscussionBoard.errors[2196110].message }}
            • Profile picture of the author jan roos
              Yeah I check how long they stay etc and another weird this is that in some niches the keyword "product name review" converts like crazy but then in a different niche the keyword converts at almost 0% but I get a ton of cheap traffic for that keyword. I guess that's why the keyword is cheap. [not converting]

              Another way I think of testing is to let a keyword at least run a full week because some days convert better than other. Sundays used to be my best days and now it's thursdays but sundays still gets the most traffic.

              To me the Adwords game is changing constantly which is why people give up fairly quick as you mentioned.

              Cheers
              Signature

              I'll teach you how to make money like a Mamba.

              Sign up for the free money mambas newsletter!

              {{ DiscussionBoard.errors[2196174].message }}
              • Profile picture of the author glenngers
                Hi - it's Glenn Livingston :-)

                The questions you're all mulling over have very much to do with the required "successful" conversion rate for your page, and the degree of risk you're willing to tolerate of (a) deleting a winner thinking it's a loser and; (b) running a loser too long while thinking it might be a winner.

                Let's say you require a 1% conversion to break even.

                Well, without going into detailed statistics, suffice it to say that it's VERY possible that your TRUE conversion rate (the one your system should stabilize at after thousands of visitors) is 1% or better, even if you had ZERO conversions after 32 visitors.

                In that particular situation, I'd be looking for at least 200 visitors. Because it's STILL possible, even likely, that your true conversion rate could be 1%+ when you haven't seen any conversions after 100 visitors.

                What if you can't afford 200 visitors to test that keyword?

                You're NOT gonna like my answer. Because if you can't afford to do the appropriate statistical test, then you shouldn't be buying the keyword in the first place. Because jumping into a market just "hoping" you'll get your conversions quickly is kind of like buying lottery tickets. Sure, you could win, but precious few people ever do.

                Now, let's say you're talking about the opt in rate instead. And let's say you know you need a 20% opt in rate to be successful.

                Well, if you had 32 visitors and ZERO opt ins, at that point I WOULD pause the system and re-evaluate. Because the odds that your TRUE opt in rate is 20%+ when none of the first 32 visitors gave you their email is very small.

                The best practice is to PLAN OUT your cash flow before you start buying traffic. How many keywords and adgroups am I going to test? How much traffic will I need before I know how I'm truly doing? Then ADJUST the number of tests you do to match the resources and risk tolerance you've got.

                I have the same "jump at it" feelings everyone else does in marketing, so I know the constant internal battle, and the above lessons were hard won with tears and sweat.

                It's the guy with the SYSTEM who wins the game.

                Hope that helps,

                Dr. G :-)

                Hyper Responsive Marketing Secrets
                {{ DiscussionBoard.errors[2202026].message }}

Trending Topics