Forum Moderators: open
Let's say you look up the term "hardware" in the goto keyword search.
Now look up "hard ware" (space).
How do you find out what people actually type in most of the time? I mean, do people look for "hardware" or "hard ware"?
The way I came upon this is that the proper spelling for
"buckskin" is one word. However, according to the goto tool, even though people bid on the word, apparently nobody has *ever* spelled it correctly when they have searched. The only spelling goto has is "buck skin" two words, which is not the incorrect way to spell the word.
Now I know that mispellings are common, but I would think there would be at least ONE search where the word was spelled correctly.
How do you find out what people actually use to search? If the goto tool doesn't tell you, what does?
-G
They not only affect words such as buckskin, smartcard, and the likes, but also everyday (there's one) words such as something, someone, therefore, whereas, everyone, and so on.
Notwithstanding the aforementioned, the only site that I am surrently aware of that has the wherewithall to give popularity information on "runonwords" vs "run on words" would be [spellweb.com...] .
Eg., Fast via SpellWeb gives 592 votes for buck skin, and 44,928 votes for buckskin.
GoTo used to give you different figures for these, I remember looking up Smart Card vs Smartcard long ago, but these have fallen foul to their spell check. Of course the real question is are they correcting correctly?
Slightly off topic, searching for references to "Run On Words" at google highlighted another problem in that google ignores the word "on" in a search for "run on words". Is there any way to search ignoring the stop word list at search engines??
Onya
Woz
The thing to remember is that the purpose of the GoTo tool is to encourage people to bid. It is not designed to be an accurate research tool. Since they no longer allow you to make separate bids on singular, plural and misspelled occurences, it doesn't make much sense for them to show bidders the counts individually. Lumping them all together inflates the overal count numbers, which gets people to bid more.
Good for GoTo, bad for us.
In Google, add a plus before the offending word.
(ex. "run +on words")
Thanks Sean
GOTO lumps together plurals, does it do this in suggestions tool and real searches? So at least they are consistant?
The tool and search results are the same, so I guess you could say they are consistant. If, for example you type in word like recipe(s), you'll only get the singular term in the keyword tool and you'll get identical search results for both variations. The same is true for misspellings
From a financial standpoint, this is probably the smartest thing GoTo has done, and IMHO probably the biggest contributing factor to their recent financial success.
BTW, a friend asked me to review a site last night. He said search on "long bow" after I specifically asked him whether it was one word or two. The correct term is "longbow" (I think).
E.g. somebody wants to search for "chess tactics" and they type in "chesstactics" ?! Or somebody wants to find "online books" and they type in "onlinebooks" ... it's bizarre.
At first I thought these people knew of some web site called "ChessTactics" or "OnlineBooks" but after studying their behavior I concluded that 99% of them are clueless.
Just like the people who don't understand the difference between typing "www.yahoo.com" into their location bar, as opposed to going to Excite and doing a search on "www.yahoo.com."
You'd think the average level of internet competance would be going up after all these years, but no... cluelessness is still king.
BOLOT
I go the other way with this Bolotomus, I think it is a sign of a maturing user, one who has the confidence to use more than two fingers to type and who makes the occasional mistake. Proof of the pudding, "nospacers" buy!
:)
I've never considered this before, but one of my clients gets a large amount traffic from a misspelling of one his keywords that is just about top on most SE's while the correct spelling I'm struggling to get in the top 5, so maybe I'll have to start looking at this strategy as well.
Yet more pennies the bankforGoto :(
Only certain words are processed similar to the "hardware" vs. "hard ware" example, i.e. the tool can only adjust terms it understands. My guess is that they process "run on words" (like Woz mentioned) and "not-really-run-on-but-big-dollar-words" (for obvious reasons).
For example, "domain names" is a popular, valuable search:
107944 "domain names"
107944 "domain name" - total combined
107944 "domainname" - total combined
107944 "domainnames" - total combined
The phrase "log files" is not nearly as popular or valuable.
639 "log files"
639 "log file" - total combined
242 "logfile" - total NOT combined
135 "logfiles" - total NOT combined
Good News: the majority of results are not adjusted when entering an "alloneword" search.
Bad News: to be useful, "alloneword" results must walk the fine line between popularity (being processed) and obscurity (zero results).
BTW - there are cases where you can make assumptions about the singular/plural popularity of the natural phrase based on the singular/plural breakdown of the unnatural phrase. (The "log files" phrase is a poor example... 1763 "nyyankees" vs. 34 "nyyankee" is a blatantly obvious example... somewhere in-between lies a more useful example.) None of this is meant to replace other methods. After all, we are talking about numbers that are manipulated directly via GoTo and indirectly via webmasters. This is just one more way to support or question keyword trends.
When all variations are lumped together in Goto, there is always the option to throw a few dollars at the Google Adwords program. They have the traffic to answer keyword questions in a hurry, and you might get a few customers in the process. (And, of course, there are the paid keyword tools, too.) HTH
contact lens - 6.2 spm (searches per million)
contactlens - 1.0 spm
contact lenses - 20.6 spm
contactlenses - 1.0 spm
That's a total of 7% nonspacers!! This blows away the #1 misspelling, which happens to be "contac lenses."
Does matchdriver on goto
only combine misspellings and variations of very popular terms?
Is there a tool that will show the difference between plural and singular search requests?
Finally, I know this is a big question but is there an easy way to compare the goto suggestion tool results with word tracker results?
Thanks
Yep, it's called Google AdWords. There was no way to decide on whether to use the run-on from GoTo data, so for a site that's getting all the product put into a directory, I asked the client to run AdWords for a couple of weeks on hand made -word- and handmade -word- and handmade is the winner in this case.
It also worked for plural or singular on another site.
It was worth checking this way before re-doing a whole product section. Simple, just choosing the one with more impressions.
1.toys 83597
2.*** toys 38813
3.Toyota 20694
4.toy 12292
5.A***t toys 12077
6.toys r us 9479
7.***toy 8033
[secure.findwhat.com...]
Keyword Center
[I'M IN THE WRONG BUSINESS]
