Why all the buzz about long tail keyword research?

Posted on March 19, 2007
Filed Under | Leave a Comment

I am terribly opinionated about the waste of time I find ‘long tail’ keyword research to be.  That’s because I look at search with a SEO (Search Engine Optimization) view not a PPC (Pay Per Click) view.  Totally different things, these two.

Pay Per Click is to Shotguns as SEO is to _____________________

What do you think the answer is?

Before you volunteer, let’s talk about pay per click and why I see it relating to shotguns.  In PPC, you pay for every visitor to your site.  Now even if you are Amazon and convert at a whopping 7%, you are still not selling 93 out of every 100 visitors.  So paying for 93 misses hurts. And if you are converting at less than 1% like most websites do, you are missing 99 out of 100 folks.  So the idea is to shoot shot at a big group of folks (a long keyword list with many different keyword phrases to target each of the individuals) and see where your shot is effective.  Then you reload as quickly as you can, perhaps modestly change your direction, and shoot again.  Conceptually as you go along, your shot becomes more effective at producing sales.  Make sense?

The biggest difference between PPC and SEO is time and specific feedback.  With PPC set up is fast (you can be live with a PPC campaign in a few minutes once you’ve done your keyword research) and you know immediately if you are hitting the target or not assuming you log into your account once a day or more.  Better yet you know which of your ’shot’ (keywords) hit the target.  So it makes sense to shoot as much shot as you can (lots of folks with different terms – not a lot of folks with one term), watch the results and then regroup and reshoot incorporating the feedback information.  Now loads of folks lose money on PPC even with all this feedback which is mostly a matter of the money loser either not accepting the feedback or ignoring the cost/profit relationship of their product market.

In SEO there is a bunch of work that starts after you complete your keyword research and then you wait and then you see if you moved forward or back and then you work again.  The time lag can be days or even weeks depending on how often a page is crawled by search engine spiders.  And if you change mulitple items at once, you have to then analyze what was the likely factor the moved you forward or backward.  Those who would recommend that anyone focus on long tail keywords for SEO is probably making the assumption that the work involved in ranking for a phrase that may get 1,000 visitors a day versus one that may get only 10 visitors a day is proportional.  Thus, they reason you should work on 100 pages for those 10 visitors each rather than work on one page for 1000.  That way if only half of them get to a top position, you are still likely to make money than to work endlessly on a competitive phrase that, they will tell you, you may not win.

The work involved with SEO ranking is not proportional.  That’s the fallacy underlying the SEO long tail theory.  Ranking for a 10 visitor a day search term is not 1% the effort that ranking for a 1000 visitor a day search term.  It’s not as much effort and it’s not as much time but it’s a lot more than 1% of the effort.

Now layer in the error rate in long tail.  What is forecasted for long tail terms can be notoriously off and you never know which terms are close to the prediction and which ones are way off.  A recent example I just learned was on a baseball site owned by a coaching client of mine.  Without any intent, we ended up ranking for ‘cheap softball gloves’, ‘cheap baseball gloves’, cheap softball bats’, and cheap baseball bats’.  These terms were pulling in a few visitors every day.  Well then I wondered how the ‘cheap’ phrases would stack up against the same phrases using the modifier ‘discount’.  I went out and checked two different keyword tools.  Here just one example:

Cheap Softball Bats                       Discount Softball Bats
Wordtracker                      22                                                 95
Trellian                            1,295                                             1085

Now just look at these numbers from a relative view because you can’t compare the gross numbers between the tools.  WordTracker says that discount softball bats is likely to be almost 5x more traffic than cheap softball bats.  Trellian says that the modifier ‘cheap’ is actually a bit more popular than ‘discount’.  So the first issue we have is who is more likely to be right on this particular query.  I highlighted that last part because I think that the one most likely to be right will change from query to query and I’ll never know which one is more accurate unless I actually successfully rank on any terms I’m comparing and check.  My guess was that Wordtracker was more accurate here.  Not sure why I thought that now.

Cheap Softball Bats                     Discount Softball Bats
Google Position                        7                                          8
MSN Position                           4                                           n/m
Yahoo Position                        n/m                                        n/m

n/m = not meaningful (any long tail position that is not on page one of a search engine is really not a meaningful position – it won’t bring any traffic)

Now here are the actual traffic results over the last 4 days:

    Cheap Softball Bats                    Discount Softball Bats
Google Traffic                        9                                                   2
MSN Traffic                           4                                                   0
Yahoo Traffic                         0                                                   0

Isn’t that interesting?  ‘Cheap’ way outperforms ‘discount’ even when we just focus on Google traffic where the positions are nearly identical.  And I didn’t put a lick of effort into getting the ‘cheap’ phrases they just came to us.  I did put work into trying to rank on ‘discount’.  I wasted my time.

Now there is one more dimension you really need to know about how SEO really works and then you’ll understand why I only spend real time on big phrases.  When a search engine finds you highly relevant on a big traffic, shorter term you will naturally get ranked on all sorts of variants were a second or third word is tacked on if all you do is place these words in your Meta tags, once or twice on your visible text and a sprinkling of times in links.  That’s really all you need to do.  Here’s an example of that.  Type in ‘golf gift’ and TGC is #1  type in ‘golf gift ideas’ and depending on the search engine, we are #1 – #3.  Now we make it really long tail and type in ‘golf gift ideas for tournaments’ and we are still between #1 and #3.  Am I trying to get ‘golf gift ideas for tournaments’?  No.  Do we have it anyway?  Yes.  That’s why long tail ranking is NOT proportional effort.  SEO doesn’t work that way.

You are better off aiming a big gun at a big target with SEO.

Pay Per Click is to Shotgun as SEO is to Cannon.

Comments

Get Your SEO News First!

Pages

• About Nancy
• Tools of the Trade
• Ask Nancy – Got a question? I’ll do my best to answer it.
• Silly SEO: Unique Valentine’s Day Gift for Grandson…

Recently

• tha action has moved
• OK – It’s True – I’m Opening Again…
• Social Media – Heaven or Hype?
• You can do this – SEO is simpler than you think
• More on How to Rank on The Search Engines
• How to Get Links to Rank High
• Is the Supplemental Index Really Dead?
• After a Too Long Absence…
• Questions about my comments on links from directories…
• Product Sourcing

Categories

Archives

• August 2008
• April 2008
• February 2008
• January 2008
• August 2007
• July 2007
• June 2007
• April 2007
• March 2007
• February 2007
• January 2007
• December 2006