The URLs That Rank And Why
View PDF | Print View
by: Vikas Malhotra
Total views: 3
Word Count: 987
We did an elaborate study of some 70 odd SEO projects in various domains: finance, real estate, retail, health care etc. We picked up all those urls that are ranking in the first page of Google.
Objective of the study, Through this study we wanted to examine the common characteristics of the urls that rank- our objective being that if we wanted to optimise other websites in the future, we could attempt to replicate the characteristics of the present ranking pages in order to achieve the same results with the new pages.
The study:
In all, a total of 272 keywords were picked for analysis We picked up a motley mix of 31websites that we were currently optimizing to find out the common characteristics of the ranking urls. Though the universal set was that of 66 websites, only 31 websites had 1 st page rankings in Google.
Assumptions:
The sample set selected was random and a fair representation of websites belonging to small business owners who do not have much Brand Equity. These business owners are primarily relying on promoting their websites for increasing their online revenue earnings through increased traffic to their websites.
For each such URL we examined the following aspects
• Ranking position
• Page Rank of ranking page
• traffic for the page
• Competition for the keyword for which the page was ranking
• Presence of H1 tag on the page with the ranking keyword
• Number of words on the ranking page
• Allinanchor position for the website for the ranking keyword
Analysis:
The trends established:
Content versus Product pages: The most prominent trend noticed in the ranking urls was the overwhelming majority of content pages that were ranking over product pages. We defined each thus:
Content pages were all those pages which had total content on the page of more than 200 words (including navigation). Product pages were those that showcased images mostly and had up to 200 words on the webpage.
We found that up to 93.75% of the ranking urls were content pages. The average number of words on such pages was 721.44 words on each page. This clearly shows that that one of the most prominent characteristics of a ranking page is its richness of content.
A steady ship needs a strong anchor
We also noticed that out of these 255 ranking urls the average anchor position was as strong as 5 when the average ranking position was 3.
There were only 61 cases of content pages having an allinanchor position for the keywords more than 10. 16 out of the above had an allinanchor position within 100. There were 45 intriguing cases of the urls rankings without a strong anchor position for the keyword.
This means that in 76 % of the cases there is a direct and strong correlation between the keyword's Google ranking and the website's related allinanchor position.
It was also noticed that the same ranking URL was not showing up for the keywords allinanchor position.
Out of the 192 urls that showed an allinanchor position ?less than 10', 177 showed the same urls as ranking in the 1 st page of Google. Only 15 cases (which is 7%) cases showed a related URL ranking for the allinanchor position.
This further emphasized the relation between good Google rankings and the importance of a strong allinanchor position for the same keyword.
When do we attempt an optimization?
Of all the keywords for which ranking could be achieved, the average competition figure was 5.6 million. The lowest competition figure was 563 and the highest (only 1 case) being 132 million. The next highest competition figure is 44 million.
What is a hopeful PR.
The average Page Rank of the ranking URLs was 3.9. 22(8% of total sample size) cases showed 0 PR. 35 (12.9 %) cases showed PR less than 3. 104 cases (38% cases) showed PR of above 5. Thus it may be concluded that while a Page with PR 3-4 can be conveniently optimized for a keyword that has about 5 million competition.
Does TLD (top level domain) matter.
Of the 272 cases studied we had been able to achieve rankings in local Google domains (.in, .au and .co.uk) in 32 cases. Of these 32 cases there were only 2 web sites which did not have a matching TLD, while 3 did. While it would be tempting to conclude that the TLD was not so important indetermining Local Domain Google ranks, it would also not be a hasty decision as the sample size in question is too small- 11.76% of the total sample size ( with only 5 sites as examples). More extensive study is required in this field in order to derive a more relevant trend.
Is there something called Ideal Keyword Density?
Of the 272 cases studied an overwhelming 194 instances were found of 0 keyword density on the ranking URL for the desired keyword! 28 cases had 1 occurrence, 10 cases had a double occurrence of the keyword, 9 cases had a triple occurrence, and 6 cases had 4 occurrences, with the ratio steadily decreasing thus. Thus it is clear that there is no direct positive correlation between keyword density and the rankings in Google for that keyword. However interestingly there were 265 (97%) cases of the keyword present on the page in separate locations (i.e.: 'car loans' present separately on the page as 'car' and ?loans'). This means that while optimizing a site for keywords, it may help to place the keywords together but the algorithm recognizes the overall number of time the words (separately) are present on the page.
About the Author
Vikas Malhotra, head of a successful SEO Services company in India, writes on Local Marketing & Web Site Development .
Rating: Not yet rated