How can we do SEO if Google rankings are never the same?

[mrez74]

< moved from [webmasterworld.com...] >

So if G is showing different results to each person then how can we go about tracking our rankings if we can only see our own results which may or may not be the same to what others are seeing?

How should we judge now how a site is ranking if our site is ranking like 50 ways if say 50 people search from 50 different states? Up to now I was under the impression that there would be slight variations depending on which datacenter you hit, but it would be minor. Like if I see my site showing at #3, then I would assume a majority would see the same (of course assuming Google.com from US, and not UK, Canada, etc.) ranking for my site with maybe 10-20% seeing at #4, #5, or small percentage at #2. Then after some time I would stabilize on all datacenters and show #3 ranking for my keyword.

Is this kind of assumption no longer valid?

[edited by: tedster at 6:40 am (utc) on May 19, 2010]

[tedster]

That assumption is definitely a goner - and it was over quite a bit before the Mayday update, too. See this discussion about rank checking tools [webmasterworld.com].

Ranking variations can come from geographic location (even within a large city), browser variety, search history when logged in AND cookie-based history when logged out, data center, and simply Google testing different variations.

[mrez74]

Tedster...if that is the case then say I start a new site and target like 25 keywords, put up the content and get it indexed. Next thing I'd like to find out is where exactly my site is ranking initially for various keywords to get a general idea of how much SEO work needs to be done.

If we are not able to at least get a rough idea of how we are "really" ranking across the board then how can we know how much work needs to be done to move up the SERPs.

Of course variations in rankings are present no matter what, but like a few spots up or down within the major datacenters and a little bit more across the rest. If I'm ranking #4 here,#15 somewhere else, #30 on another datacenter, etc.. all the time then this becomes a real issue.

I would think that if you built a site with great on-site and off-site SEO (along with Google trust)then your rankings should not be all over the place compared to other sites. From this simple testing it seems that at least the #1 site is the only consistent site for pretty much everyone.

[walkman]

Location definitely changes serps and so does 'history'

[freejung]

The simple answer is that our primary metric is now traffic, not ranking.

[aparraga]

yes, I have a main keyword that ranks #4 in spain but #1 in USA, and I assume it because of the location.

There are so many parameters that G takes into account for that (the location of your server, if your domain is a .com, .us, .es, ..., the language on the page, if any address on the page (e.g. in the footer), if any currency symbol on the page, where are who link to your page from (also where are who you are linking to), the language of the anchors (or even the pages) that link to you, etc...)

The list never ends, however, the important thing here is that it's natural to have different results in the serp depending on your location.

[cangoou]

There's never a reason to wait on improving content, getting good backlinks...

The Q is, if you where ranking on page 1 and you are now lets say page 3 or 4, is it:
- Just a Google hickup (so building new links is fine)
- Some devaluating in ranking-factors (so building new links is fine)
- A kind of penalty for building too much links (so building more links could make it worse)
- A kind of other penalty (so perhaps you should change your IBL-Profile, even perhaps deleting some links)
I always find it hard to say which one you should go for...

[mrez74]

Cangoou...my issue is that when I do a search for my keyword the last week or so it shows my site ranking at #3 or #4 compared to #10-#12 a few weeks ago.

Now I'm not seeing any improvement in traffic from moving up from the bottom of page 1/top of page 2 to the current position at #3/#4. I know there is traffic for this keyword from my previous experience. But when I check WMT for the same time period it shows an "Avg Rank" of 10 for that same keyword that I keep showing #3/#4 for the last week or so on the actual SERPs.

So from that it seems that on most of the datacenters that people search for I'm still not showing up that high, but much lower (compared to my search position when not being logged in,cleared cache, on desktop, laptop,Iphone, etc).

So at this point I'm still building backlinks and adding more content. Hopefully I starting moving higher on the rest of the datacenters and start to see more traffic.

This difference between what I see and what WMT shows as average rank is present in several of my sites. However, for a site that I rank #1 now for like 2 weeks WMT does show an Avg Rank of "1.4" and the impressions/clicks pretty much are inline to what I'm seeing (compared to statcounter). So I guess from now on I need to rely on WMT Avg Rank to get an idea of how my keywords are doing and not totally rely on the rankings I get from my end.

[dickbaker]

I don't trust the rankings shown in WMT any more than I trust the number of links or other data.

I've checked various datacenters and I don't see my keywords or phrases ranking as low as WMT indicates they can be. For example, WMT may show the CTR for a phrase in the #2, #8 or page 2 spots, but the phrase is always in the top five across all the datacenters.

On the flip side, WMT will show a great ranking for a keyword or phrase that's never ranked well no matter which datacenter I've used.

[maximillianos]

You really don't need to analyze rankings. I stopped checking my rankings a year or two ago. Mostly because I get traffic from a half million key phrase combinations a month, but also because it fluctuated so much and didn't seem to matter in my traffic.

What you should spend your time on is analyzing your traffic, what pages are folks coming in on, what keywords are they searching for to hit those pages. Why are those pages a success and others not? What can you do to optimize the pages that are not getting much traffic... etc.

If you try to worry about rankings, you will end up going crazy and have many sleepless nights. Focus on what you can control and see in your own data and feedback from your users.

[rach251]

Yeah I'm of the opinion that rank checking, used as the primary method of measuring search engine performance, has pretty much gone the way of the dinosaurs. I still do it just to check for major problems but if I had to use it as my primary metric I would go nuts.