| |||||||||
![]() |
|
|
«
Previous Thread
|
Next Thread
»
|
Thread Tools | Search this Thread | Rate Thread | Display Modes |
|
#1
|
||||
|
||||
|
I have been attempting to find a good method to estimate traffic based on numbers solely from Overture & Wordtracker - without having to rely on setting up a test PPC campaign. In order to do this, I used 4 websites which rank #1 for 4 different terms at the 3 major engines. I then engineered a formula using Overture & WordTracker's numbers to come up with a system that would fit the facts.
1st - I combined all of Wordtracker's "predict" numbers for plurals, capitalization, etc. for the target term into a single daily prediction number. 2nd - I took Overture's numbers and divided them by the number of days in the month (for this example it was February, so I used 2). These two numbers I then plugged into Excel to try to find a formula that would fit all 4 of my data points. I shortly came up with something that did: Avg Traffic/Day = (0.5*OvertureNumber)+(0.5*WordTrackerNumber)*0.015 This formula produced numbers that were +/- 15% of my actual traffic levels for every term for which I have a site/page ranking #1 in Yahoo!, MSN & Google... 37 terms in total. It's not scientific, and I don't represent it to be flawless, but I am going to try using this formula in the future to estimate traffic, until I find it to be significantly off the mark. Some clear inaccuracies include: 1. Good vs. Poor Title & Description in SERPs 2. Seasonal Variations 3. Weekly fluctuations (more/less traffic on weekends) 4. Polysemic terms - those that have multiple meanings I'm hoping that as a community, we can try this out and see how many people find it to fit their data as well, and how many find it bogus. Thanks for your input! |
|
#2
|
||||
|
||||
|
Hey Randfish, I'd love to try this out, but I don't have any wordtracker numbers to work with. Anything that can be done just using overture that would be semi close to your combined results?
Thanks for working hard on this! Cheerios
__________________
Cheerios! New to SEO? See the FAQ! My Disclaimer: Don't Listen To Me - I know nothing! |
|
#3
|
||||
|
||||
|
That's pretty damned cool Randfish!
G-Man
__________________
Black Hat, White Hat and More! |
|
#4
|
||||
|
||||
|
Rand:
Let me restate this. You seem to be saying that you are getting about a 1.5% click through rate of the DAILY calculation of 1/2 of the combined totals for Wordtracker and Overture. Just a word of caution. Daily rates are subject to significant fluctuation. Weekly and then monthly rates will even out the daily figures. You will be subject to economic conditions, possibly seasonality (possibly not) and competition. Wonder who else has tried to calculate site clickthroughs versus traffic volume. Actually Googles numbers from adsense are our best numbers. Since google delivers about 60% of our current search traffic (even with some #1's in Y and MSN) it is the telling information as to traffic and effectiveness. This is where EGOL really leads the way, experimenting with his presence on the web and continuously testing to get better volume and better conversions. It is valuable information. Post 2/2/05 we are facing competition on the web for our site/business and we have to get better. Good effort. Dave |
|
#5
|
||||
|
||||
|
Quote:
Randfish: You really need to put up adsense. It takes the guess work out of this for google traffic. Currently we are facing a level of web competition that we haven't seen in the last 1 and 1/2 years. Now we have to get much better and tougher faster. That's alright we have faced competition for over 20 years. Anyways adsense stats will give you actual google impressions on various keywords. It provides tighter and better numbers than the wordtracker and overture estimates. For us google delivers about 60% of our traffic even as we have a number of #1's in MSN and Yahoo. I've got to go back through EGOL's lessons. That dude works over his keywords, positioning and traffic like a master. Dave |
|
#6
|
||||
|
||||
|
gchaney - You can get Wordtracker numbers for free by signing up for the trial or using Digitalpoint's kw tool.
Dave - I agree, this is a flawed method. I'm just wondering whether it is accurate "enough" to be used as a guesstimate before launching an AdSense campaign and Overture campaign to really check the numbers... I'm just wondering if my formula fits other people's numbers as it fits mine. Remember that these aren't straight OV or WT numbers - they've been modified, OV by dividing it into days and WT by combining all capitalization and plurals. I think both are inaccurate, but I'm seeking a temporary solution that will be useful in a preliminary estimation phase. |
![]() |
| Viewing: SEO Chat Forums > Search Engine Strategies > Keywords > Evaluation of KW Traffic Estimate Formula |
| Thread Tools | Search this Thread |
| Display Modes | Rate This Thread |
|
|
|
|
|