| 8:38 pm on Dec 4, 2008 (gmt 0)|
Hi Halim :)
As an aside, we avoid the use of specific keywords if possible - widget being a good replacement ;)
If Google is your target, then it's an easy test:
These types of searches generally return:
- Different results
- Some sites in both sets of results
- Some sites that appear in one set of results, but not the other
So, the suggestion is:
- It's possible to rank well for both
- Some sites won't rank for both
- Google does not consider them to be identical searches, with identical relevance
- You don't need to be using both in your textual copy to appear for both variations (although, of course, most sites do)
Google is IMO some distance ahead of it's competitors in terms of "translating" a search keyword so as to retrieve documents that don't necessarily contain the exact wording the searcher used (query expansion, as it's known). So you can optimise for word variations without particularly trying to do so. But equally (especially if the keywords are competitive) you want to back a winning horse, in case you don't succeed in appearing for both.
Your other consideration is whether plural or singular forms are better at driving the right visitors. For transactional keywords there can be a big difference in both traffic and conversion rate for word variations. The goal is to predict searcher behaviour. So is "Buy a widget" better than "Buy widgets"? It depends on both the conversion rate and overall traffic volume as to which is better for a particular site.
Adwords is a great playground for keyword research, if you're prepared to spend money to get reliable data. IMO it can be money well spent if you're making decisions with long-lasting implications. Otherwise, there are SERPs and free keyword tools on which to base any decisions.
| 10:29 pm on Dec 4, 2008 (gmt 0)|
Thank you Andy for your quick response.
I won't use specific keywords on my next posting...
| 11:38 am on Dec 5, 2008 (gmt 0)|
>>Google is IMO some distance ahead of it's competitors in terms of "translating" a search keyword so as to retrieve documents that don't necessarily contain the exact wording the searcher used <<
Does the same apply for grammatically incorrect keywords or misspelled keywords (as adwords sometimes suggests)?
| 11:50 am on Dec 5, 2008 (gmt 0)|
Yes - all can be mapped to an occurrence of another, related word or phrase. I did start a thread about query expansion and Google [webmasterworld.com] a while ago, which may be of interest.
They are still different keywords, with different relevance, but it is possible for a single item of content to appear for many variations, without using them on the page itself or having incoming links with those related words.
| 12:09 pm on Dec 5, 2008 (gmt 0)|
So if I have a page of content how can I be sure what keywords it's ranking for then?
| 12:14 pm on Dec 5, 2008 (gmt 0)|
It's impossible to know which words and phrases resulted in your content getting an impression in the SERPs - only Google know that (they provide scant data about it in WMT). Your web analytics will tell you whether keyword-targeting is successful, and if you are targeting particular words and phrases you should be checking the SERPs themselves.
| 12:20 pm on Dec 5, 2008 (gmt 0)|
I see. Thank you.
on a slightly different note though when choosing keywords to target for is it better to take a product oriented or market oriented approach?
| 12:29 pm on Dec 5, 2008 (gmt 0)|
That would depend on an individual market, and is probably a question better suited to it's own thread (so we don't hijack Halim's topic!). Good keyword research should help to get a reliable answer for your own market.
| 2:26 pm on Dec 5, 2008 (gmt 0)|
I use Google quite a bit in researching various topics for a discussion forum, and it's always amazed me how Google will return a different list of hits for the singular and plural forms of the same word. And yet, despite seeming to be very particular in that respect, it's also amazed me how Google will return a very similar list of hits for two words, one of which is spelled correctly, and the other is not, and even if you really hash the spelling it will offer the correct spelling. e.g. "widget" and "widgets" return a different set of hits, whereas "widget" and "wijit" might return the same or very similar set of hits, and when you use "wijit", Google will ask "did you mean widget?"
| 6:55 pm on Dec 5, 2008 (gmt 0)|
The majority of search engines today handle singular and plural keywords differently. Were they to treat singular and keywords the same, the search results would be the same, evidently they are not.
Try it now by searching the major search engines for both <a singular keyword and a plural keyword>, the results are often remarkably different:
It is important therefore, to optimise a website for both singular and plural keywords and in proportion to their respective usages. If most people would search using a plural than more effort should be directed towards optimising the website for the search phrase involving the plural keyword.
[edited by: Receptional_Andy at 7:14 pm (utc) on Dec. 5, 2008]
[edit reason] No specific keywords, please [/edit]