When Not To Optimise, When the tail is longer than the head Options

[fisicx]

I'm seeing more and more searches that don't fit into the 'normal' keyword pattern. It may be that people have got used to asking questions rather than just typing keywords into the search box or it may be that they have always done so and the search engines have changed the way they deal with such queries.

I've started this topic in response to the What Is Seo These Days? thread. It is my contention that optimising for specific keywords is no longer a core activity. I have a page which has the title and main header 'fixing browser problems'. It ranks well for this phrase but nobody lands on this page for those keywords. This is a selection from the last couple of days:

javascript resize web page to browser screen
website screen resolution problem
wide screen scroll problem ie
fix browser problems
make browser +wide +screen
1024x768 javascript resize browser
how to fix all monitor resolution in all browsers
ie7 javascript resizeto
ie7 resize browser window
1024 javascript resizeto ie7
fit browser to screen size

As you can see, there no consistency in the search pattern. If I had 'optimised' the page for MY keywords I may have lost all those visitors and THEIR keywords.

So here is the question: how important is keyword research when writing content?

<stop press> This just in, a search for: 'can you use dropdown menus for screen readers and use text for mobile users'. Don't get much longer than that.</stop press>

This post has been edited by fisicx: Jun 18 2008, 03:33 AM

[A.N.Onym]

As you can see, the keywords are related to the article, so I don't see a problem. If you optimized for some phrase or simply added it to the title (Fixing Browser and Resolution Problems), then you'd get more traffic on the same article, if you didn't change the text.

One issue that might make you research your keywords is to actually optimize for the phrase that people actually search for. If you don't research and end up having your URL/title/anchor text focused on the wrong keyword, you won't get much traffic, except the very long tail.

The reason you don't show up for 'browser problems' (not in top 100) is because your page doesn't have many links pointing at it, so you are left with the long tail.

If you do keyword research, it has been my conviction as of late to focus on medium/high competitive keywords and write lengthier posts. The long tail would attach itself easily.

That being said, another area of keyword research is discovery of topics to cover (such as harris tweed jacket for men, jacket hood insulated men and so forth). You wouldn't uncover these topics without either an extensive guide on jackets (and including the above by luck and even more extensive keyword log analysis) or simply using tools to learn about these subjects people are interested in.

This post has been edited by A.N.Onym: Jun 18 2008, 05:33 AM

[EGOL]

If you optimized for some phrase or simply added it to the title (Fixing Browser and Resolution Problems), then you'd get more traffic on the same article, if you didn't change the text.

I agree.

I like these kinds of query problems. There is an enormous diversity of search for this and related topics. What I do with a page like this is run weblogexpert and see what WORDS (not phrases) pull the traffic. Then I do two things...

1) Think about new and relevant search terms that people might use to land on this page

2) Lengthen the article, going into a lot more detail. (this will qualify you for a lot more long tail search queries and you can get an enormous amount of traffic from them - sometimes a lot of that traffic will be relevant)

The approach above is not something that I would use for a retail site to pull traffic with a high probability of conversion although it might improve sales. This is a traffic gunner method.

[fisicx]

I agree!

The point of my orignal post was not how to get more traffic but to question the need to optimise for specific keywords. Yes, put together a page title/description/header package but I'm finding more and more that it is the long tail that is bringing in the traffic. People no longer carry out one or two word searches, they type questions, select multiple terms, use delineators, use alternative words, spell it wrong, use acronyms and abbreviations.

Egol - you are right, I'm already adding to the article as I discover what peole are looking for which in turn is bringing in more traffic.

Yuri - If 'browser problems' was a popular search term then I would agree - but it isn't, people don't do this search - they search instead for the solution to a specific browser problem (I've done my research).

I'm not suggesting that keyword research is dead - but I am suggesting that you should ensure that the page contents contain a lot more material than just focussing on the keywords you have chosen.

(I'll change my page title and see what a difference this makes).

[EGOL]

I think that the most important optimization for pages like this is the title tag...

...instead of writing it with word selections that pull traffic, write it with word selections that pull clicks. If you can do both that is even better.

[dazzlindonna]

Let's face it. You gotta put "something" in that title tag, or in that H1 tag. It might as well be a keyword phrase that is relevant to the page and that is the preferred phrase to rank well for. Just having long tail phrases throughout your content will enable them to rank generally, so you pretty much don't have to do much optimizing for them anyway. Just.write.words.

So, I don't see any reason to avoid optimizing for good phrases, when the long tail terms usually take care of themselves as long as you've actually used them somewhere in the copy.

Both can play well together at the same time.

[fisicx]

I AGREE!

I've been advicating the title tag and description for pulling in the visitors for yonks. But I still believe that the way people search for information is reducing the importance of ranking well for just one or two keywords.

The content defines the keywords but the title can only relfect a very small part of those keywords. My arguement is that it is not worth worrying obout your rank for specific keywords if the long tail is pulling a high volume of traffic. As Egol suggests, build up the content to increase the long tail traffic.

I beleive it's what Ammon was indicating here

This post has been edited by fisicx: Jun 18 2008, 09:58 AM

[ukdaz]

People no longer carry out one or two word searches

Really? Not from where I am sitting.

Yes some users are becoming more savvy but I work on site A LOT with my clients and when I observe them using a search engine the vast majority still only put in two or three word searches. I've yet to see any client - use a search engine in any other manner.

Maybe its just me?

This post has been edited by ukdaz: Jun 18 2008, 12:56 PM

[dazzlindonna]

It's not just you.

I make a fairly large amount of my income from people who have searched for two word phrases. I'd lose a TON of money if I ignored those phrases.

In my niches: lots of people are still using important, generic, two word phrases.

[Black_Knight]

Actually, you now have to optimise for both things. The head and the tail.

To optimise for the tail, you use underlying SEO, as in a solid foundation of good SE friendly code, plus great content that exercises a good vocabulary and covers the subject from a variety of angles, preferably offering paths to related content with differing focuses too. You build up the site's authority through both solid backlinks, and through cultivation of co-citation with relevant resources.

Across a wide spectrum of client sites, I've found it a good indication of getting it right when around 50% of traffic is coming via the phrases we optimise for, and 50% come from long tail search. That's because that's about the split in actual searching I believe.

So, if the large majority of your traffic is pure long tail, then you are missing the 50% of search use that's using only 2-3 keywords, and quite predictable phrases. You don't rank #1 for "fixing IE" or "internet help" I'm sure. I'm sure, because if you did so, then you'd see the 'head' terms still drive plenty of traffic.

That said, it does vary a lot by topic once you get commercial. There are market sectors where it is essential to capture the 'head' traffic, because that's the branding space. You need to catch head terms when you are pitching an unusual or alternative offering. Before the searcher on those early forays to see what the broad range of options are misses the new option you're presenting for the traditional solutions. In such cases, one may have a page, or several pages that seriously target only head terms.

Most homepages only tend to attract relatively 'head' type traffic. There are exceptions, but if you look at all search referrals that land at the homepage on almost any well-optimised ecommerce site, its the shorter, more predictable 'head' terms that drive >80% of the search traffic to that page.

Like almost everything else in SEO, there's a balance to be struck, and the optimal optimisation is invariably a compromise of some sort.

[A.N.Onym]

Yuri - If 'browser problems' was a popular search term then I would agree - but it isn't, people don't do this search - they search instead for the solution to a specific browser problem (I've done my research).

Precisely one of the points of keyword research: you'd rather optimize for something people search for, not something they don't.

In each case of search query numbers and content, there are different choices. As usual, it depends.

What it all comes down to, though, is optimizing for both, as Ammon has astonishingly well elaborated on.

This post has been edited by A.N.Onym: Jun 18 2008, 10:48 PM

[fisicx]

Thanks for the excellent advice. I'm off to challenge the dragon and see if I can't boost my visitors using 'two keyword searches'.

You are all of course quite right, I have been missing the point - just because people don't find my site using one/two keyword searches doesn't means it's not happening.

Thank you all - I am humbled.

[Black_Knight]

I am humbled.

On the contrary, my old mate, I think you've expanded yourself again.

Of course, the real challenge is going to be finding the 'head' terms that people are using. There's really no easy way to get those terms, and instead it is going to take creativity, and a lot of testing. The easiest place to start however is to look at the head-style referrals you get even where you're only in position 20+ in the SERPs. If those terms are driving traffic even from the third page (which only around 12% of searchers ever look at) then there's obviously a lot more mileage in them. (Plus if people are drilling that far, no site before is fully answering the question, right? The old supply and demand situation that indicates a good market condition).

Other than that, you can be using things like the Alexa traffic ranking to spot similar resources with a lot more traffic, then looking to work out their best performing 'head' phrases. Backward engineering the competition. There's a couple of online resources that help you spy out the keywords that are working for others which can help too.

Finally, don't forget PPC. There's a good chance that the short and relevant phrases that ther sites are bidding most for are the terms they've found perform best for them. Using the PPC keyword tools can help you draw upon the experience and testing of others.

[Joshua Sciarrino]

Across a wide spectrum of client sites, I've found it a good indication of getting it right when around 50% of traffic is coming via the phrases we optimise for, and 50% come from long tail search. That's because that's about the split in actual searching I believe.

Wow, I never heard that before, I guess it comes with the territory of being a newbie to SEO.

Black_Knight, can you explain this for me?

To optimise for the tail, you use underlying SEO

What do you mean 'underlying SEO'? Thanks in advance.

[Black_Knight]

Black_Knight, can you explain this for me?
What do you mean 'underlying SEO'? Thanks in advance.

I partly explained it in the first post:

a solid foundation of good SE friendly code, plus great content that exercises a good vocabulary and covers the subject from a variety of angles, preferably offering paths to related content with differing focuses too. You build up the site's authority through both solid backlinks, and through cultivation of co-citation with relevant resources.

Underlying SEO is the strength that a page will have just from going up onto the site and in its content, before specifically optimising that page. The power of the site itself is a large part of that, through factors like its link popularity, its structure, its status as a hub/authority, and so forth.

This post has been edited by Black_Knight: Jun 19 2008, 11:47 AM

[Joshua Sciarrino]

I partly explained it in the first post:

I was thinking you did. Thanks for the clarification.