How do you determine keyword profitability?

Discussion in 'General Marketing' started by AfterHim.com, May 28, 2005.

  1. #1
    I have all of my stats...

    # of searches
    # of sites in G
    amount paid by advertisers of adsense

    Can someone smarter than me help me figure out a keyworld profitability index of sorts?
     
    AfterHim.com, May 28, 2005 IP
  2. Design Agent

    Design Agent Peon

    Messages:
    3,061
    Likes Received:
    154
    Best Answers:
    0
    Trophy Points:
    0
    #2
    I would say there are too many variables. First, I am assuming you mean adsense, not adwords.
    Second, What are you trying to work out exactly - having a list wont help as payout is dependent on lots of factors bid price, CTR, ad relevance to your visitors etc
     
    Design Agent, May 28, 2005 IP
  3. AfterHim.com

    AfterHim.com Peon

    Messages:
    1,923
    Likes Received:
    51
    Best Answers:
    0
    Trophy Points:
    0
    #3
    DA,

    I meant, what you mean. I meant adwords, and so did you ;)

    I am just trying to figure out a little system for myself...right now it is pretty random, but I am looking to see what other people do as well (if anything).

    Brandon
     
    AfterHim.com, May 28, 2005 IP
  4. AfterHim.com

    AfterHim.com Peon

    Messages:
    1,923
    Likes Received:
    51
    Best Answers:
    0
    Trophy Points:
    0
    #4
    Anyone?

    On a side note, I'm pretty sick of getting negative votes for no reason on threads like this...
     
    AfterHim.com, May 28, 2005 IP
    howardroark likes this.
  5. howardroark

    howardroark Active Member

    Messages:
    127
    Likes Received:
    1
    Best Answers:
    0
    Trophy Points:
    53
    #5
    i think the yahoo marketing tools are the best way to determine which keywords have the highest payout (publishers) and which keywords are most searched often (advertisers). although i noticed that google adwords and adsense are often higher (because they are more popular i guess)

    i wouldn't mind the negative votes. if you look at the ones with high votes here, many of them have useless posts and talk about things unrelated to web dev or marketing, anyway. there, i gave you a positive vote :)
     
    howardroark, May 28, 2005 IP
  6. Dominic

    Dominic Well-Known Member

    Messages:
    1,725
    Likes Received:
    121
    Best Answers:
    0
    Trophy Points:
    185
    #6
    If you can get a distinct segment of a demographic you can earn more from working individually with advertisers that want access to that segment.... just something else to think about.

    Otherwise, I think you are on track with your other indicators. Next I guess is to determine how to monetise the ranking.
     
    Dominic, May 29, 2005 IP
  7. virginia-home

    virginia-home Peon

    Messages:
    29
    Likes Received:
    1
    Best Answers:
    0
    Trophy Points:
    0
    #7
    Thank you for your note!
     
    virginia-home, May 29, 2005 IP
  8. AfterHim.com

    AfterHim.com Peon

    Messages:
    1,923
    Likes Received:
    51
    Best Answers:
    0
    Trophy Points:
    0
    #8
    Dominic,

    I was actually looking for a formula...I think I'll try to come up with one myself..unless someone has one and wants to pm me :)

    brandon
     
    AfterHim.com, May 29, 2005 IP
  9. datadugout

    datadugout Peon

    Messages:
    60
    Likes Received:
    0
    Best Answers:
    0
    Trophy Points:
    0
    #9
    Hi Brandon...

    I don't really have a formula, but I look at many things....

    1) The keyword tool at Overture to start

    2) I also look and see how many pages Google has indexed when I type in the search term (with and without quotes). Same thing for Yahoo..

    3) How much the main PPC's (AdWords) are getting from advertisers. Check out http://www.googspy.com

    4) I also look at eBay and see if there is competition there for the keyword. If so, generally there is some money in it. of course it also helps to see if anyone is bidding on the item!

    3) If things look good so far I then look and see who is on page 1 of the SERPs for Google and Yahoo for my main keyword and check out their site and see how easy it will be for me to beat 'em. ;)

    Hope that is somewhat helpful.

    -datadugout
     
    datadugout, May 29, 2005 IP
  10. compar

    compar Peon

    Messages:
    2,705
    Likes Received:
    169
    Best Answers:
    0
    Trophy Points:
    0
    #10
    I've read the entire thread and I don't understand the question. What is keyword profitability?

    Are you talking about a $ figure? That's how profitability is normally measured?

    Are you talking about a word's potential to get your site in the SERPs?

    Are you talking about determining the optimal price to bid in either AdWord or Overture -- or whatever it is called now?
     
    compar, May 29, 2005 IP
  11. DarrenC

    DarrenC Peon

    Messages:
    3,386
    Likes Received:
    154
    Best Answers:
    0
    Trophy Points:
    0
    #11
    I'm with Compar with this.. it's like asking how long is that peice of string.

    If your referring to how do you find out if the keyword you are optimising your site is a popular keyword which potentially could bring profitability, then in my experience it's a lot of trial and error.

    You can use tools to give you an estimated popularity of the keyword but it's a case of seeing and then evaluating what the popularity of the keyword is from your own stats and income.

    Darren
     
    DarrenC, May 30, 2005 IP
  12. tradefor

    tradefor Peon

    Messages:
    98
    Likes Received:
    7
    Best Answers:
    0
    Trophy Points:
    0
    #12
    I have to think it would be possible to come up with a formula for estimating keyword profitability based on optimizing for the various obvioius parameters mentioned here: keyword search frequency; the level of $$ bids for keywords on Overture and AdWords; CTR and CPM for relevant AdWords ads carried as AdSense ads on your site.
    The problems I see with that are:
    1. Basing any optimized formula on a very limited sample. I don't know of an archive of historic data for any of the variables, except of course your own logs; so that you may be stuck with drawing inferences from only one or two months data which may not be representative; e.g., the ad metrics for snow-chains in January and February, as compared to projecting demand for July and August.
    2. You may end up over-optimizing/ over-fitting the formula for your product only, and because of (1) above not be able to test it on out-of-sample data, although you could if you cared to test it on a completely unrelated product, albeit only for the most recent month's worth of data.

    One caveat: I am new at this, and I don't use Overture myself; so I may not be aware of some archive of historical search frequency that would make for a valid sample for formula testing purposes.

    There was thread a few weeks ago that described a bunch of tools assembled by GoLexa.
    http://forums.digitalpoint.com/showthread.php?t=13921&highlight=Golexa
    including stuff from Nichebot. Judging by the results for Business Cards
    http://www.nichebot.com/?PHPSESSID=d59d108ad7062e8cc9e4c68707e04890&term=Business+Cards
    you are all over the "low competition key phrase balanced with relatively high search frequency" idea already!!
     
    tradefor, May 30, 2005 IP