~~~High Search Engine Rankings~~~
From: GFX Printing
Hi Jill and Heather,
In #050, Heather, when responding to Michael about "tshirts," although
she may not realize it, brings up a new question (I checked the
archives and didn't see it raised): is now having singular and plural
variations important?
In the olden days when I first began submitting my own site to search
engines (c. 1995), I believe it was Altavista that said not to worry
about singular and plural usage because a search for "postcards" would
show all results with the word "postcard" as well (that was true until
at least 1998 I'm sure).
These days, it doesn't appear true at all ... doing a search on
Google, for example, will result in wildly different listings for both
the singular and plural version of the same phrase.
In the last couple of years, I guess I've been a little busy and
haven't noticed such a drastic change, and was wondering if this
relates to the majority of search engines and directories now.
If so, it certainly does make writing more challenging, e.g., postcard
printing is grammatically correct, but postcards printing sounds silly
:) (I know I could just replace a few instances to printed postcards
to allow for plural variations ... but I guess it means a lot of work
ahead of me, if indeed this is now true).
Thanks muchly,
Barb
GFX Printing Services
~~~Jill's Response~~~
Hey Barb,
Thanks for bringing up an interesting subject that we haven't yet
tackled. You'd think that after a year of Rank Write we'd be running
out of new topics for discussion, but you guys keep just 'em coming!
WTG!
I must say that it has *always* been my understanding that it was
important to use both singular and plural variations of your
keyphrases if you wanted to rank high for both forms. I felt that way
back in '95, and continue to feel that way now. However, I haven't
given it much thought lately, and figured now would be a good time to
do a little bit of research. (Thanks a lot, Barb for making me do
some actual *work* for this newsletter...I'll bill ya later!)
I remember hearing or reading about some engines' use of "stemming"
which I knew had to do with plurals and singulars, etc., so I did a
bit of research on that first. I found a quickie explanation of stemming at Search Engine Watch. Danny
Sullivan says, "Stemming is the ability for a search engine to search
for variations of a word based on its stem. For example, entering
'swim' might also find 'swims' and maybe 'swimming,' depending on the
search engine." Apparently Lycos and NorthernLight use this stemming
feature (or at least they did when Danny's article was written at the
end of last year). Many other engines have a stemming feature you can
turn on by using their Advanced search.
Thus said, just because an engine *can* find the plural when a
singular form (or vice versa) is typed in the search box, it doesn't
mean that it *will* find it, or will find it first. If all else is
equal, chances are that the engine will first find the matches that
correspond with the exact search term, and then find the other forms.
At least that's my assumption. So before making the proverbial ass
out of myself by ASSuming stuff, I spent my morning doing some
detective work.
Here's what I did, along with the results of my highly UNscientific
study:
Using our "tshirt" example from last week, I went to each major search
engine and did a search for "tshirt" in one browser window, and
"tshirts" in another browser window. In nearly every engine,
different results came up for the singular and the plural searches,
plus I found some other interesting and unexpected surprises along the
way! Because of these surprises, I also did a search for "women's
tshirts" and "women's tshirt" so I could compare the results of a
keyphrase with that of a keyword.
In AltaVista, the results were basically as I expected them to be.
The search for "tshirt" brought up sites that used the singular form
of the word in its Title tag, and the search for "tshirts" brought up
sites with the plural form in its Title tag. This was fairly
consistent throughout the first two pages of results. (I didn't go
much further than that on most engines.) Along with sites that had
these words in the Title tag, there were a few strange and apparently
irrelevant sites also mixed in with the results. There didn't seem to
be much variance with the two-word phrase of "women's tshirt(s)"
either.
Google showed the same type of phenomenon, only with much more
relevant results (sorry AltaVista but you've got some major catching
up to do!). Again, the Title tag seemed to have a strong relevance
with what showed up in the listings. As with AltaVista, there didn't
seem to be much variance with the two-word phrase of "women's
tshirt(s)" either.
At HotBot, I saw a slightly different story. The first page of
results was very similar for either "tshirt" or "tshirts" (not exact,
but similar). But after the first page, the results began differing.
They appeared to place a lot of emphasis on the Meta description tag
and that corresponded with whether or not I searched in singular or
plural terms. (Please note that I did not actually click on most of
the sites and did not look at the text on the page, but based my
findings on the information that showed up in the list of results.)
So why was the first page of results at HotBot inconsistent with the
next page of results? I believe this is due to the fact that Inktomi
and/or HotBot have some human editors that actually handpick certain
sites for some competitive search terms. I remember having read or
heard somewhere that they do this, and it does appear to be true.
Which also might explain why it's so difficult to get a top-ten
listing for some keywords on HotBot these days. Now, using the
two-word phrase, I didn't get a sense that the first page of results
was "fixed." They were the same crappy results that you tend to get
most of the time at this engine these days. Apparently HotBot hasn't
heard of the Title tag and its relevancy to page content!
Lycos had completely different results for the singular and the plural
forms of "tshirt." One thing that was consistent was that many of
their results appeared to be totally irrelevant to my search terms.
Shame on Lycos. The results for the plural form appeared to be
slightly more relevant than those for the singular form. But many of
them still sucked, for the most part! I also did a quick check of the
Fast database at AllTheWeb and got the exact same results that I got
at Lycos. So perhaps we can simply blame Fast for the poor show of
relevancy. The search for the two-word phrases gave slightly better
results. Interestingly enough, Lycos' "popular" results showed the
same sites when using "women's tshirt" or "women's tshirts" but their
"web" results showed different sites. With the one-word keyword, the
popular results showed differing results. Not sure what to make of
this, but it's interesting, nonetheless!
Next was Excite's turn. Beats me if anyone searches there anymore,
but I figured I should check it out anyway. Interestingly enough, the
results for both the plural and the singular forms of "tshirt" were
*exactly* the same! That was a bit of a surprise to me. It appeared
that they were probably searching for the plural form, for the most
part. Even more surprising was that when using the two-word phrase,
the results were suddenly *different* for the plural and singular
searches. But wait, it gets even better. The searches for both forms
of "women's tshirt(s)" at Excite brought up mostly porn-type sites
(wet tshirt kind of sites)! This did not happen in any other engine.
So if you're looking for porn, Excite may be your best bet.
(Heather...take note!)
I also checked out NorthernLight just for the heck of it. Man, did I
get some crazy results from them! The plural and the singular brought
very different results, but all were stuffed with numerous seemingly
irrelevant pages. I also noticed that Titles didn't seem to play all
that much of a role in what would come up at NorthernLight. Strange.
And even stranger was the fact that the two-word phrase brought up the
same exact results whether it was plural or singular. So different
results on the one-word keywords, but the same results on the keyword
phrase. I'm sure there's a good explanation for this, but I can't
quite fathom what it might be at the moment!
So what does all this craziness mean to poor Barb and her long
forgotten question? Well, in my opinion, it's back to my original
thought. As I've always advocated, if you want to be found under both
forms of the word, use both the plural and the singular forms in the
Title and the other tags (as well as in your copy). So yes, Barb, you
may have your work cut out for you. But then again, perhaps not. It
might be enough for you to simply add in the singular form to your
Title tag if it's not currently there. You might give that a try for
a few months and see how it goes.
My testing has also shown me that I've been right all along about
shooting for keyphrases as opposed to keywords. The results for
single keywords are much more irrelevant in most engines. There's no
sense optimizing for one word if the engines can't categorize one-word
keywords properly due to their vagueness. So keep on looking for the
most appropriate keyPHRASES and you'll have a much better shot at high
rankings AND targeted traffic!
And the last thing that my testing showed me (which I already knew),
is that Google totally dominates as far as the relevancy of search
results goes. I don't know what in the world has happened to
AltaVista and Inktomi. I distinctly remember that at some point in
their pasts, both of these "engines" had fairly relevant results (the
rest of them were never very relevant!). Who knows what they've done
to mess it up, but mess it up they have! This leaves me to wonder
just how long most search engines will be around. As more and more
people become Google converts, why would they ever go back to any
other engine? From looking at my traffic logs, it already appears
that searches (from real people) must be waaaaaayyyyyy down everywhere
but at Google, Yahoo!, AOL and MSN. The bulk of my sites are getting
most of their traffic from those four engines/directories, even when
they have just as many high rankings on the other search engines.
Goodbye other search engines - we knew you well...
Jill
~~~Send Us Your Questions~~~
If you have questions about online copywriting or search engine optimization (or both!), just zip us an email to questions@rankwrite.com. We've had some folks ask if their question
was "too basic" to be printed - and you don't have to worry about
that! There are no "stupid" search engine optimization or copywriting
questions, so ask away!
