Astonishing Facts about Keywords and Traffic

[GeorgR.]

I made a quite lengthy post about this here, with pie-chart etc..

A Shocking Truth about Traffic to your Web Site, SEO and your Keywords

In my opinion, this is really, really astonishing.

* Over 50% of website traffic comes from what i want to call "one-shot" keywords which "basically do not exist" (until someone uses this phrase on Google)

To confirm this i just added a recent Google analytics shot to demonstrate the significance of this to your SEO and keyword research.

* A month's worth of traffic came from about 2000 keywords.

The "best" keyword in itself is only responsible for 15% of ALL traffic, and the keywords 1-20 etc. only for a little less than 50%.

The rest, responsible for 51% of all traffic, is keywords which had 1 (ONE!) search and hit in this month.

Keywords which simply "do not exist" (because they do not have search volume) and no SEO software, adwords keyword tool etc. will show them, ever.

Interesting stuff indeed...

[Kurt]

Originally Posted by GeorgR.

I made a quite lengthy post about this here, with pie-chart etc..

A Shocking Truth about Traffic to your Web Site, SEO and your Keywords

In my opinion, this is really, really astonishing.

* Over 50% of website traffic comes from what i want to call "one-shot" keywords which "basically do not exist" (until someone uses this phrase on Google)

To confirm this i just added a recent Google analytics shot to demonstrate the significance of this to your SEO and keyword research.

* A month's worth of traffic came from about 2000 keywords.

The "best" keyword in itself is only responsible for 15% of ALL traffic, and the keywords 1-20 etc. only for a little less than 50%.

The rest, responsible for 51% of all traffic, is keywords which had 1 (ONE!) search and hit in this month.

Keywords which simply "do not exist" (because they do not have search volume) and no SEO software, adwords keyword tool etc. will show them, ever.

Interesting stuff indeed...

I've posted time and time again since I've been a member here (9 years) that about half of all the traffic SEs send are from "one of a kind"/unique search queries that the engines don't have any real results for.

And IMO, Google treats these types of searches a little differently than the more popular queries. I believe that these searches have always relied more upon On Page and very little with linking.

The key to exploiting this strategy is to use as many keywords in as many combinations as possible, as opposed to the traditional advice concering focusing a page's SEO on a single keyword phrase.


I made a thread here about how people really search:
The Most Important SEO Lesson of the Week

On the thread I posted a text file that contains ACTUAL searches done in real time (at that time) when this data was available. The point is, people search in crazy ways and their queries aren't as neat and tidy as our keyword tools suggest...And for about half the queries, Google doesn't have a clue what to show in the SERPs.

[TryBPO]

This is sort-of old news. The length of search queries has been increasing over the years, as mentioned briefly in this article in 2008: Long Tail Searches Are Becoming More Mainstream

There's another interesting point from the guys over at Search Engine Land about whether instant search will kill the long-tail. You can read that here: Will Google Instant Kill The Long Tail?

[GeorgR.]

While many of those phrases ARE indeed long-tail, they are not necessarily all. Sometimes its just "odd" search phrases you didn't think of before.

What struck me most is looking at the stats like bounce rate and page views/visitor...which shows that the visitors coming from the "one shot" keywords spend significantly more time on the site and check out way more pages - as opposed to the traffic from the "main" keywords.

[TryBPO]

Originally Posted by GeorgR.

What struck me most is looking at the stats like bounce rate and page views/visitor...which shows that the visitors coming from the "one shot" keywords spend significantly more time on the site and check out way more pages - as opposed to the traffic from the "main" keywords.

Hmmm...that is interesting. Did you find that across the board? Will check our site and see if we find the same thing...

[Kurt]

Originally Posted by TryBPO

This is sort-of old news. The length of search queries has been increasing over the years, as mentioned briefly in this article in 2008: Long Tail Searches Are Becoming More Mainstream

There's another interesting point from the guys over at Search Engine Land about whether instant search will kill the long-tail. You can read that here: Will Google Instant Kill The Long Tail?

Actually, these really aren't long tail keywords, they are unique, one-time search phrases...It's doubtful optimizing pages for them will bring any future traffic like typical long tail keywords might.

You're correct about the length of search queries getting longer...They have been since 1996, when I first started SEO. Google instant could affect the one-time search queries, if for no other reason than simply distracting searchers and is something we'll have have to keep an eye on.

[GeorgR.]

You can't optimize for those phrases because you cannot "predict" the phrases or what people will be looking for.

Also..it wouldn't make sense to optimize for a keyword which has "0 searches" ..respective which is a one-shot wonder.

The only (?) good strategy here is to have lots of good content to increase the chances that such unique keywords are appearing on Google for many searches.

If you have a site consisting of only one or two articles..chances are slimmer than if you have a content rich site.

[Kurt]

Originally Posted by GeorgR.

You can't optimize for those phrases because you cannot "predict" the phrases or what people will be looking for.

Also..it wouldn't make sense to optimize for a keyword which has 0 searches because..respective which is a one-shot wonder.

The only (?) good strategy here is to have lots of good content to increase the chances that such unique keywords are appearing on Google for many searches.

If you have a site consisting of only one or two articles..chances are slimmer than if you have a content rich site.

You could optimize for them after you get a visitor for one by checking the logs as you did, but there's no point, which was my point.

Technically, the search total would be at least one, regardless of what the research tools tell us.

I've used a few strategies over the years, one obvious strategy is the use of spinners to get as many different keyword combinations as possible.

[TryBPO]

Originally Posted by Kurt

Actually, these really aren't long tail keywords, they are unique, one-time search phrases...It's doubtful optimizing pages for them will bring any future traffic like typical long tail keywords might.

You're correct about the length of search queries getting longer...They have been since 1996, when I first started SEO. Google instant could affect the one-time search queries, if for no other reason than simply distracting searchers and is something we'll have have to keep an eye on.

How would you not consider those long tail keywords? Just because there was only one visit doesn't mean there was only one search performed that month. There might have been 2, 4, 5, etc. searches and only one visit...the site owner wouldn't know that.

Either way, I would agree that it's at the LONG end of the long tail phrases and definitely not worth optimizing for. As another poster said, those will come naturally with your content anyway.

[markowe]

This sort of goes back to Google's mantra about writing good, useful content, bla-di-bla. There is SOME truth to that, I guess I am trying an experiment where I actually write some really helpful useful content (yeah, call me crazy) - I mean REALLY good, Wikipedia-beating stuff in the niche. I am finding an amazing range of long-tails (one-shot jobs, like you say) bringing visitors to my site. Oh, and that's with no real off-site SEO.

This has reminded me to go and put WordPress › SEO SearchTerms Tagging 2 « WordPress Plugins on that site, and claim those terms for good. Who knows, someone MIGHT search for that again one day

[merlincat]

Since writing for my Kitchen accessories site i've noticed loads of searches coming in on words I would never have thought of, some of them are one off and if I start to see a trend, such as Turquoise kitchen accessories then I write a page just to target these keywords.

But yes I'd have to say that over half of my visitors come from words I haven't optimized for, and then they stick around for a while.

So I think it's a combination of targetted seo and the lucky break that lots of content just happens to bring.

[Kurt]

Originally Posted by TryBPO

How would you not consider those long tail keywords? Just because there was only one visit doesn't mean there was only one search performed that month. There might have been 2, 4, 5, etc. searches and only one visit...the site owner wouldn't know that.

Either way, I would agree that it's at the LONG end of the long tail phrases and definitely not worth optimizing for. As another poster said, those will come naturally with your content anyway.

I explained my reasoning for giving these keywords a different definition than long tail, as you said, a site owner wouldn't know any different.

However, my definition isn't based on what a site owner knows or doesn't know. It's based on the concept that it's estimated that about 50% off all search queries are unique and result in only one search. And if they are unique and only produce one search, they can't be searched for 2, 4, or 5 times.

IMO, the concept and strategy is important enough that these types of keywords shouldn't be confused with "long tail". Which BTW, is not a term I ever really liked...

As a matter of fact, a few people on this thread have verified that they get half their traffic from these types of keywords. The problem becomes that people assume that 50% is "OK" and comes naturally. However, with proper SEO this number can be increased, which is what optimization is.

If you want to depend on these keywords "coming naturally" to your writing, that's your decision. I do something else, which is to think of ways to use as many keywords as possible, while making it APPEAR to look "natural", so I can get as many combinations of keywords as possible and vary things like combinatins, proximities, order of the words, prominence, etc, which are all important SEO factors.

[FrankLinks]

I think it's not easy to make many keywords to appear to look natural in one article.is there any software can help us to do this thing?

[brobdingnagian]

Great thread. Thanks!