11 replies to this topic

[1dmf]

I read with interest Jill's article on rankings being a poor measure of success and I understand alot of the reasonings behind this.

But was wondering if perhaps it wasn't as clear cut as that?

I guess the main question is 'where do most websites obtain their traffic'. and is the answer 'Search Engines'?

If so , then ranking has to matter just as much as it used to surely?, the problem is it's a lot harder to get a good ranking for every single searcher, due to localisation / personalised search etc. etc.

Or do most visitors now come from vistor streams and no longer from search engines? be it links on popular websites, paid link adverts, social media and other internet marketing?

If rankings are no longer important, then that would mean we no longer need to link build, as boosting one's PR is irrelivant as it doesn't matter where you rank in the SE's.

Or should that guide you as to how you link build, why you link build and where you link build?

I can certainly understand Jills comment

Rankings don't equal targeted traffic. Heck, rankings don't always even equal un-targeted traffic! If you or your SEO company optimizes your pages for keyword phrases that nobody's searching for, your optimization efforts will all be wasted. And if you're measuring success by how you rank for those useless keywords, you may be thinking you're successful when you're really not. This is actually one of the oldest tricks in the book for unscrupulous (or incompetent) SEO companies to use. They fulfill their end of the bargain—get you rankings—and you're left scratching your head wondering why your website is still a ghost town.

I often wonder why for some, what I would call 'generic' keywords relative to my page, I rank 1st page and receive little traffic.

How do you know if it's a poor keyword, good keyword but little traffic anyway or little exposure due to poor 'Global' rankings. Even if you rank high when you search for yourself!

I appreciate everyones input on this.

1DMF.

Edited by Jill, 01 December 2009 - 01:47 PM.

[Jill]

Of course you need to be showing up in Google when people are searching using the words related to what you off, for which you've optimized.

The problem with rankings is that people think that they're number 3 or number 10, when in reality, it's possibly just what they happen to be seeing at that given time/location/machine.

My Pubcon presentation illustrated this pretty well. I did some searches on my computer at home using Firefox where I was also logged into my gmail account. Screen capped the results, then did on same computer using an incognito Chrome browser, not logged in, and the results were different.

Then did same searches in Vegas from my laptop and yet again, different results. Now, they were similar, in often the same 20 sites were still the same 20 sites, but mixed up in positions. But which site was #1 and which was #4? You can't say because it's going to be different for everyone.

This is the reason why I say rankings are a poor measure of success. I get calls all the time where they say I'm #5 but really want/need to be #1. 'cept they may not be #5 always and to everyone.

That's why the GRE tool is so nice because it tells us exactly what rank we were when we got an actual click. And 1dmf, surely you've seen from your own results with that tool that you see the same search queries at different positions at times?

That said, you can have certain authoritative articles on a subject that do indeed stick pretty well in their coveted spots. But even those waver slightly at times, I've noticed this via the tool as well.

[qwerty]

I don't think rankings are completely useless, but I don't think of a good ranking as a goal. It's more of a potential explanation of why a given page is or isn't performing well. Let's say a given site regularly gets 50 visits a month from a particular search, and that number changes drastically one month. I'll take a look at the ranking to see if that might be the reason (or part of the reason) for the change.

But if you come at it with your goal being a good ranking for a keyword, you're missing the point, because a high rank doesn't necessarily translate into traffic or conversions. Besides, you can't really report on rankings if they're different for every user.

[chrishirst]

Ranking is useful as that is where SE referrers come from. It's the checking of "rankings" that is the useless bit. What you see is not what other searchers and there fore leads you to erroneous conclusions.

[1dmf]

So it's as I thought, about who see's your ranking! and doing what's necessary to get more people seeing you highly ranked!

So the SE principle and main optimisation hasn't changed, just our ability to see 'real' results through searching ourselves.

As you say Jill, the GRE tool, is certainly going to give a much better overview of ranking, once G! rolls it out across all it's data centres.

Once more ranking positions become available, I think and 'average rank' would be a handy little tid bit of information, what you think?

[qwerty]

Once more ranking positions become available, I think and 'average rank' would be a handy little tid bit of information, what you think?

If it can be shown that that average rank brings in traffic, if that traffic increases with a higher average ranking, and if that traffic can be made to convert, then yeah. Otherwise, it just provides you with bragging rights.

[1dmf]

Well unfortunately the GRE Tool is never going to be able to help you convert traffic, it only records it!

[Randy]

Actually, it could be integrated into a conversion tool 1dmf. Just that I have no plans to do so myself.

That said, I may contact the guy who created the <cough>little</cough> conversion testing tool I use and see if he feels like integrating the GRE way of revealing rank into the tool. He's already capturing some other stuff, so adding in Google rankings for those with the special URL shouldn't be a big job for him, and would improve data on the back end for those searches. Both for converting and non-converting traffic.

The issue there of course being that only Google provides those special URLs. And only on about 1/5 of all Google searches.

[1dmf]

True, in fact didn't you mention you were going to add conversion stats into the GRE tool? Have you changed your mind?

It would require the tool to do far more sophisticated tracking of visits that's for sure!

Supplying the average rank will also require a lot of server side processing if just wacked onto the current table view, especially if it's got to be grouped by search phrase.

hmm, might give me something to do over the Crimbo holidays!

[Jill]

I wouldn't go to too much trouble, guys. I assume that eventually Google will add the rankings into GoAn at some point anyway, making GRE obsolete.

[1dmf]

Osolete, never!

Title Edit -> Rankings Are Dead, Long Live GRE!!!!

[Randy]

I didn't do full blown conversion tracking in my tests 1dmf. I simply integrated a new database field in my normal (php based) conversion tracking software when appropriate. It already grabs keyword data, so I had to do was tweak its code a bit using GRE logic to have it look for the cd= string and record in the conversion tracking database too. That's why I think Jim might add it in a new version of his software. It's really a minor hack once you wrap your head around the concept.

Conversion tracking back to the original keyword phrase and ranking could be done fully inside GRE also. It would just take a bit of IP tracking and/or a cookie on the front end during the initial hit then an extra database field to place an entry in when someone reaches the confirmation page. Pretty simple really.