How to target longtail 3 or more keyword groups etc?

[Mark_A]

So, I am not having as much success with two word key terms, partly because they are popular and the competition seems quite entrenched, partly because our sites just don't have enough inward links and or PR for these queries.

However I notice that for some longer key-term searches we do well, way up on page 1.

So I want to identify important 3 and 4 key-terms and go after them looking for rankings and click throughs.

Anyone have any thoughts about the mechanics of actually doing this?

[tedster]

Study the Google Sugggestions, for starters. Suggestions are heavily dependent on search volumes, so you can get a good idea about what related phrases the average person is searching on.

Don't just spam every phrase you see there - that can get you penalized. But when you see suggestions that fall squarely within your area of focus, then make sure you've got extended content focused on those phrases.

[goodroi]

from my pov there are two big issues - identifying & building.

not every 3 or 4 word keyword combination is going to make your money. use google suggestions, adwords keyword suggestions, your log files, your competition log files (amazing how many people store them online & unsecured). these sources should identify keywords with decent traffic. a bit of trial and error should then help you figure out which ones with traffic actually turn into money.

as for building out, i have used two main strategies. for the really good keyword combinations i register a domain & build out a minisite with real content. for the rest of them i build out several pages of content for each keyword phrase and interlink them with decent anchor text. in my experience building a single page generally is not going to be enough.

[Mark_A]

Hi tedster, thanks for your response. I don't think I have ever used the google suggestions tool. No time like now to try though. What I noticed was that in WMT a search for "red and round widgets" had ocurred 170 times last month and that we had an average position 6 which is good but we got less than 10 clickthroughs.

Interestingly the exact phrase does not appear on our website, and in only one of the top ten sites does it appear exactly like that at all. Plus there was a mix of high and low PR sites returned for that search.

I figure if I can work out other multiple word searches that are relevant for us and then include the term just once on the site, we might get to P1.

[Mark_A]

Hi goodroi, yes, identifying may be tricky, if we do not get returned for the term then it will not show in WMT even though it may be both relevant for us and be being searched for.

I think building minisites is not on for us, we want the main site to remain the centre of our online marketing. And not sure how many pages of content I can build. I was hoping to customise existing pages by inclusion of the multi-word key phrase and the like.

Looking at WMT's view of the keywords on our sites is fascinating, the most popular keywords on our site are quite good (i.e. what we want) but they are also competitive and as such hard to rank for, even for two word combinations.

[piatkow]

Long tail searches can be weird. I have seen my site at around number 4 when it has the exact phrase and the three above it don't.

The first thing is to have the relevant words in the content. The mistake that I made with a new site was to write the content as if it was a print publication. While there are phrases that can be omitted in print because they do not add to understanding they can be critical for searches.

For a web site tautology is good.

[RP_Joe]

"I want to identify important 3 and 4 key-terms"

What I do is watch the stats carefully. When I see a good 3 word phrase, I check to seek what the ranking is for that. Then If I don't have a page already, I will make a make for that phrase. Keyword1-keyword2-keyword3.html . Then depending on how much value I think those KW's are I will further optimize the page. LTKW's have less traffic so you need more of them. Which means you normally can't spend allot of time on them.

[tedster]

For a web site tautology is good.

And used well, it makes for more engaging writing that is comprehensible to a wider audience. With the long tail, it's not so much that you directly target a whole bunch of combinations - but more that you make your pages available to those queries. And the traffic may be lower volume, but it tends to be very well targeted.

Taken in aggregate this traffic can be a very effective way to build your online business. Over time, it can help you achieve higher rankings for mid-range phrases and then some real power terms.

[Planet13]

LTKW's have less traffic so you need more of them. Which means you normally can't spend allot of time on them.

On the other hand, they MIGHT have a higher conversion rate, or be a big revenue generator (even with a lower conversion rate). So check on how much money those particular pages generate and see about linking to them directly from the home page so as to boost their PR.

[Hoople]

I did a single page for a local dance that had to 'compete' against the 800 pound gorilla's in the room, dating sites.

For a web site tautology is good.

Saying who it was for in many ways (men, guys, gentlemen) helped greatly. Like RP_Joe I also watched the stats and added words in search terms that lead them to the page.

[Mark_A]

Thing is..

Using WMT or whatever I often only see the key-terms for which my site is returned.

I really want to see how many searches all relevant key-terms are getting.

I used to get that using wordtracker but I have not used it for yonks!

[Sgt_Kickaxe]

Just write content for people, forget Google.

[piatkow]

Just write content for people, forget Google.

Up to a point.
Write content for people, yes, but not in the same style as you would use for print. Google is still an important secondary audience which cannot be ignored. As I said in my previous post:

While there are phrases that can be omitted in print because they do not add to understanding they can be critical for searches.

[RP_Joe]

Planet13, your are correct, they on average do convert better.

A recent statistic I saw was that 55% of all searches were 3 words or more. 70% have no exact match. So there is a big opportunity. All of my pages now are 3 words or more.

.

[Mark_A]

Just write content for people, forget Google.

It is quite possible to write for people but no people find the page at all. I believe in writing for the dual audience:

1) to attract SE rankings to get people viewing
2) to be interesting or informative for people..

[Mark_A]

I just had 7 related but different three word terms and have managed to use them all reasonably well in two shortish paragraphs!

Go chew on that google ! :-)

[wingslevel]

the longtails have been a hot topic over the last 6 weeks around here. i think they are especially tricky because they are often pretty far down the tree branches of site architecture - example 'frozen fuzzy blue widgets' - to get there on an ecomm site you might have to drill something like this: home> widgets> frozen> colors> blue - prolly not likely to be a whole lot of tbpr showing on that page - so now you put 'frozen fuzzy blue widgets' in the page title, description, url and maybe an alt text and - whammo - you trip some kind of oop filter so you are -950'd or worse - thats cause google is seeing a highly optimized page without a lot of link support - but you just might end up #1 for 'online fuzzy frozen widgets' - go figure