| 4:54 am on Nov 7, 2007 (gmt 0)|
|I've always taken the approach that the most important things for good rankings in the future would be the amount of time people spent on my site, the number of page views, a low bounce rate and a high book marking rate. |
Your list is missing backlinks. Especially for rankings on Google, backlinks is a very big factor, the monster in the corner.
|SE - Bounce Rate |
google - 71.37%
yahoo - 39.85%
altavista - 21.43%
msn - 11.11%
Now those are some might suspicious numbers. First thing I would do is break it down by search phrase, not just a global trafficmeasure. I would also break that down further by the specific landing page
My first take is that your competition would not be able to affect your site very much this way, if at all. But who knows? That's why digging deeper makes sense to me. You may be ranked on Google for phrases that poorly target your pages. Other patterns may emerge as well.
| 10:43 am on Nov 7, 2007 (gmt 0)|
I understand the importance of backlinks. I have a few articles out there as well as a Y directory listing, quite a few other directory listings as well and some natural links. I rank fairly well in Yahoo for some pretty competitive terms, just nothing in G.
The terms I'm talking about are Title terms I use in Google base. I've tracked enough of them to know they're coming directly from Base. It's not that hard since I don't get any organic G traffic anyway. The terms are something like "Blue widgets by Mr X". That term would take you directly to the page titles "Blue Widgets by Mr X" and the widgets are right there. It's not like they're looking for information, they know it's a product to purchase when they click on it, it is after all Google Base.
The most telling sign to me is the "0" time spent. That means that over half are clicking on a link marked "Blue Widgets" and than closing the browser before the page even loads. And believe me, my pages load fast. I think I'm being played :(
| 12:00 pm on Nov 7, 2007 (gmt 0)|
Do you have any images on your pages? An attractive image indexed in Google images can produce very high bounce rates.
| 2:10 pm on Nov 7, 2007 (gmt 0)|
Yea, I do get a lot of image searches but those aren't the ones causing the high bounce rate. I've tracked a good deal of them by search term and I'm sure it's 99% Google Base which is why I believe it's a competitor.
| 2:30 pm on Nov 7, 2007 (gmt 0)|
|Since I first built I've always I've always taken the approach that the most important things for good rankings in the future would be the amount of time people spent on my site, the number of page views, a low bounce rate and a high book marking rate. |
Makes sense, and its been talked about quite a bit, but exactly how did this become universally accepted? What real evidence does anyone have that G, or other search engines, use these metrics to rank sites?
| 3:33 pm on Nov 7, 2007 (gmt 0)|
You said it yourself you get no traffic from google so the number coming is much much smaller thus the overall average is outta wack.
You get organic traffic from the rest so that number is a much larger number and a more accurate number to deal with.
A competitior since you don't have any google listings wouldn't waste their time to bounce in and out besides the IP is collected so it will only count once not multiple times.
There is another issue with the site it isn't a competitor and since you said from the start Google has had an issue with your site it may just be worth you time to have the site examined by others here. Not sure if you are a member but it would be well worth the amount spent to get some of the best in the business take a look.
[webmasterworld.com...] Have to be a member but like I said if I didn't have a good answer as to why I have always had an issue with the Google algo I would pay for a review...
I have nothing to do with this site just a member myself but it looks like you are grasping for anything as to an answer to the Google problem were it may be a simple fix or a complex answer... Let the members help you find the issue.
| 4:29 pm on Nov 7, 2007 (gmt 0)|
"What real evidence does anyone have that G, or other search engines, use these metrics to rank sites?"
None, it just makes sense. If not now, than in the future.
"it may just be worth you time to have the site examined by others here"
Thanks BWN, I think that's exactly what I'll do. But I still think something is not right. The numbers could be skewed slightly but those numbers are based on over 500 clicks from Google Base from the past week. No matter how far I go back the numbers are the same. To have over HALF come back with "0" time spent and 100% bounce?
| 5:26 pm on Nov 7, 2007 (gmt 0)|
I think you're misunderstanding some analytic terms.
If you have 100% bounce rate on a page then time on site MUST be 0. A bounce is when a visitor viewed 1 page and then left. An exit is a when a visitor viewed at least 1 previous page and then left on the page you're seeing the stats for.
| 5:27 pm on Nov 7, 2007 (gmt 0)|
Just pointing out that Google Base is not Google and cannot be compared to conventional "search engine" traffic. I suspect most responders are talking about Google's main search engine rather than Google Base.
As for whether the bounce rate can affect your ranking there, who knows? I just did a search for one fairly common term, and I got page after page of book listings all from the same publisher's website. That search tool has got serious problems.
I understand you want to hold on to the traffic you're currently getting, but IMO you'd be better off reading tutorials and looking at your site through spider emulators, in order to figure out why Google's main search engine doesn't send you anything.
| 5:34 pm on Nov 7, 2007 (gmt 0)|
have you done a search in Google products and then gone to the link. Are you sure the correct page is being loaded.
That is strange.
Before you submit your site please go to at least two other sites and give some input on thier site ok. it is a requirment to keep the board.
| 5:41 pm on Nov 7, 2007 (gmt 0)|
FWIW I list products in Google Base and they convert like crazy. However it's not the Base listings that convert (honestly who the heck uses that?) it's the google base product listing in the normal google results where google lists 3 products and price at the top of the page.
| 6:41 pm on Nov 7, 2007 (gmt 0)|
Jomaxx has it right. Anything with one page view show's zero time spent. This would make sense since people using base are generally looking for one thing in particular and not wasting time surfing. I just hope G takes that into consideration because my stats look terrible because of the lack of natural listings.
| 6:13 am on Nov 8, 2007 (gmt 0)|
But if we ignore Google Base your original question is a good one that deserves more attention.
Google is always striving to deliver better search results. They can track either through analytics (if you're using it) or through their own results page how long a surfer spends on a page. If visitors are only spending 2 seconds at the #1 result page then it probably isn't a great result for that keyword.
In theory it sounds like a good way to systematically judge how well search results are for a keyword but programatically it would be too easy to manipulate.
It could easily be manipulated by yourself (by using scripts that fake long visits to your site) or by your competition (by using scripts that always bounce).
[edited by: BradleyT at 6:13 am (utc) on Nov. 8, 2007]
| 12:41 am on Nov 10, 2007 (gmt 0)|
|Since I first built I've always I've always taken the approach that the most important things for good rankings in the future would be the amount of time people spent on my site, the number of page views, a low bounce rate |
I strongly believe these are myths.
Here is why.
Time on site:
Some of the most valuable sites to me are the ones I stay on the LEAST amount of time, because I got my information FAST.
When I can get the information I am looking for on the page G finds most relevant, voila! No need to go any further! Awesome!
So if G were to award sites that I am STUCK on longer and have to CLICK on more pages, that would SUCK.
I think your assumptions are off.
I even doubt clicks on the results page matter.
Just look at the verticals. I saw Yahoo present at SES and they said their objective is to "give the information the user is looking for on the search results page itself."
That's why when you search for a pizzeria in NYC, you get the numbers, locations, distance, and names on the results page.
Think about it: the search engines want people to stay on their engine to keep searching and creating more ad inventory.