Is Long Tail a Myth?

[Tonearm]

Has anyone else seen astoundingly little from the long tail? I read about it a long time ago and added a lot of extra keywords to my site in hopes of pulling in long tail searches. I just realized I've seen little to none in all this time. Maybe you have to be dealing with extremely high volume stuff to see it produce?

[crobb305]

No myth. I have always received a significant amount of traffic from longtail (4+ words), and I consistently ranked top 10 for single/two-word phrases last year. Glancing at my stats for 2009, it looks like about 40% to 50% of my searches were from phrases of 4+ words. I think it helps if fragments of those longtails appear somewhere in your backlink profile (not necessarily exact match, and not limited to on-page). Furthermore, I find that my longtail phrases convert better.

[maximillianos]

How many pages do you have on your site? Adding "long tail" key phrases to, say, 100 pages isn't going to change much in regards to your traffic if those pages are already optimized for regular keywords. You need 100 new pages that are optimized to the new longtail key phrases.

The cases I've seen that are most successful with longtail are the ones where sites have hundreds of thousands of pages of unique content. Then you have a situation where a longtail page brings in 4-5 visitors a day, and if you have 100,000 of those pages, you are looking at close to a half-million visitors a day... you get the point. It is all about scaling it up.

[crobb305]

I definitely agree, the more pages the better. My site is 107 pages, and was receiving 3500 to 4000 uniques a day, 80% entering via the homepage. I don't have the staff to do sites with thousands of pages. I have to make-do with my limited resources. Longtail was definitely a winner for me though over the past 4 years. (I say "was" because I am looking at 2009 stats collectively. I have not had time to look at 2010, but 2009 gives me a good sample).

[youfoundjake]

I'm crushing .edu and .gov sites in my niche with longtail searches. Too bad the ratio is something like 10000/1, but I'll take it.

[martinibuster]

Go to Google Trends and type in one and two word keyword phrases and review the history of it's use. In most cases you will see a trend of less people using those phrases.

Then go to Google External tool and type in those same one and two word phrases. Select "exact match" so that it returns results for those exact phrases, unmodified by other keyword phrases and take note how many queries per month it reports. For the next step select Broad Match and review how many more searches are found for the keyword phrases modified by other phrases (like color, stores, and words such as cheap and on-sale). In most cases the broad match contains more keyword phrases. This is the longtail.

[Tonearm]

Thanks for the insight guys. Why is creating more pages so important for the long tail? What's wrong with adding keywords to pages you already have?

[kevsta]

its definitely not a myth but longtail volume is dependent on the size of the site and increases exponentially with size.

we have a site with approx 20k pages, top kw term (non brand) is doing about 4k a month, by kw 20 on the list it's 100 a month, by kw 100 on the list you're at 5 a month, by kw 200 it's 2 per month, and then there's 65k more one off hits below that per month.

at least 60% of the total traffic could be described as longtail.

[maximillianos]

Thanks for the insight guys. Why is creating more pages so important for the long tail? What's wrong with adding keywords to pages you already have?

Just my opinion, but if your pages are already ranking well for certain keyword searches, that implies they are optimized well for those keywords. Optimizing the same pages for additional long-tail phrases might end up hurting your existing traffic, and cause a wash in the end...

Think of it this way, one page can't be about everything.

[wheel]

Has anyone else seen astoundingly little from the long tail?

A few years ago I added thousands of pages of longtail to a site. Traffic doubled. Not all of it as targetted as I'd like, but it was there.

I just increased my pages from abut 10 to 5000, Google's indexed about half of those. I haven't measured the long tail (working on it right now) but we are getting a few telephone calls per week on stuff that is clearly coming from our long tail.

[martinibuster]

What kinds of longtails are there?

Off the top of my head, there are:

• shopping longtail phrases
  
• how-to, when-to, and what-to longtail phrases
  
• highly specific longtail phrases (ISBN, colors, zip codes, area codes)
  
• phrases composed as questions
  
• location specific phrases
  

Throw in misspellings and the list grows exponentially. :P

[Tonearm]

Just my opinion, but if your pages are already ranking well for certain keyword searches, that implies they are optimized well for those keywords. Optimizing the same pages for additional long-tail phrases might end up hurting your existing traffic, and cause a wash in the end...

Think of it this way, one page can't be about everything.

Is it a keyword density issue?

I just increased my pages from abut 10 to 5000

Can anyone give me an example of how something like this might be done? What are these extra pages about relative to the 10 main pages?

How can this type of thing be accomplished on an ecommerce site? I have category pages, sub-category pages, and product pages. I can add new categories, sub-categories, and products, but that isn't really long tail, right?

[wheel]

I have category pages, sub-category pages, and product pages. I can add new categories, sub-categories, and products, but that isn't really long tail, right?

Maybe, maybe not.

I've got a dell 4700C as my office computer. I want some more ram so rather than crawling under my desk I search on 'dell 4700c memory'. That's a long tail search term. If I land on your shopping page that specifically discusses my options and prices for Ram for my PC, you've greatly increased your chances of making a sale. Ranking on 'memory' or 'ram' or 'cheap ram prices' won't put you in front of me when I'm searching on long tail stuff - I didn't even search on that term.

[Tonearm]

I've got a dell 4700C as my office computer. I want some more ram so rather than crawling under my desk I search on 'dell 4700c memory'. That's a long tail search term. If I land on your shopping page that specifically discusses my options and prices for Ram for my PC, you've greatly increased your chances of making a sale. Ranking on 'memory' or 'ram' or 'cheap ram prices' won't put you in front of me when I'm searching on long tail stuff - I didn't even search on that term.

Nice example.

And it's not possible to effectively optimize your RAM page for 'memory', 'ram', 'cheap ram prices', and 'dell 4700c memory' because you can't keep the keyword density high enough for each term?

I can imagine building a whole slew of pages linked from my RAM page, each targeting very specific keywords, but then you've got so many links on the RAM page that you can't pass enough PR juice to each one to keep them in the main index.

[martinibuster]

>>>because you can't keep the keyword density high enough for each term?...

That may not necessarily have to be a consideration for longtail optimization. These phrases are so unique that very often there aren't many other web pages that can match all the elements in the query. The traditional approach of "targeting" doesn't need to be followed as rigorously. Like the difference between a single fishing line for catching a single fish and casting a wide net.

[BillyS]

Yesterday's statistics, 1,200 page website:

25K visitors via 14K keywords / key phrases - it's not a myth.

Do we target two word phrases? Absolutely - but the competition is extreme and we're up against BIG players. We will never win that battle. BUT, I don't believe that our success with long tail phrases is luck either. Maybe we’re kidding ourselves; maybe we found the key to this success. Not sure, only search engines know the answer. I do believe there is a writing technique (or formula) that leads to this success.

When I retire, I’m going to write a book on this topic. Unfortunately, this is one of the very few bits of information that I won’t share on this forum. I’ll also never divulge how much I make (or don’t make) publicly either.

I’m not doing this to tease anyone, only to share a success story on this topic. I’m not here to make enemies either (although I got a nasty bit of hate mail recently from a member of this forum). I want everyone to feel the success I’ve experienced – so I’m still here. Like me, or hate me.

Sorry for the digression. Long tail is real.

[tedster]

Absolutely, Billy. People who create pages with only minimal amounts of text never get to see long tail traffic - because there's not enough variety on the page to even GET a long tail hit from. But as you point out, if you aim for a high volume phrase with enough supporting content, then the first place you can see search traffic is often from those longer phrases that branch out from the trophy phrase.

If you mine that long tail keyword information from your server logs or analytics, you can see where you are currently rank - and then work on improving those rankings, especially for long tail terms that are more than just one-off occurrences. If the URL starts to get enough clicks from related long tail (or, I suspect, even just enough impressions in the SERPs), then you can see the trophy phrase ranking start to improve, too.

It's a fascinating process to work with - and you're mostly building from real traffic to begin with, not theoretical traffic that a keyword tool seems to promise. Not that keyword tools research is a bad idea - it really helps, too. The whole thing works together.