September 2002 Volume VI Issue Nine Entire Issue

�

Link Popularity vs. Link Reputation – you decide
by Leslie Rohde

Terminology varies somewhat depending on who you talk to, but generally Link Popularity is a count of links to a page, whereas Link Reputation involves some form of analysis of the text of links that point to a page – aka, incoming text-links.

There are a number of tools and web sites that provide measures of Link Popularity and ways to compare the popularities of competing pages. These tools do have some limited use in finding linking partners. The question is, however, are they useful in obtaining higher search engine ranking?

Does the page with the most links rank higher? Or is it the quality of inbound links that dictates ranking? If Link Popularity really does lead to high search engine placement as some people claim, then we should see the top pages consistently having higher link counts. But is this the case? Let’s take a look.

The top ten sites of 2.7 million at Google for the term hats have link counts as follows:

We can immediately see that something other than Link Popularity is at work here.

• The top page (which ends up at both #1 and #2) does have the most links, but it appears to arrive there by “brute force” as we’ll see in a minute.
• The #4 page, with only 38 links must be doing something very right, to beat out pages with 16 times as many links.
• And the #9 page with only 46 links is, comparatively speaking, also doing very well in spite of a relatively low incoming link count.

In any case, it’s obvious the notion that just increasing the link count will lead to higher rankings is not entirely true. And, if you’ve ever created or obtained 900+ links then you know what a challenging task that can be! So, …what works the easy way and why did those pages with so few links score so well?

The answer is found when we analyze those pages in a manner similar to the way search engines analyze them. True, counting links is part of the equation. However, an even more important part of the formula involves…

1. taking into account the incoming link text,
2. comparing it to the text of the page it links to and
3. matching this text to the user’s search query.

This more sophisticated analysis is called the Link Reputation approach.

By analyzing the top six listings from the table above in terms of Link Reputation, we get the following data…

Let’s work from the bottom up.

• The #6 page has the second highest link count of the group yet scores the lowest in spite of having a desirable directory listing at ODP.Upon further analysis (using OptiLink) we discovered a very common mistake that needlessly “penalized” this page in the rankings – specifically 356 of those 626 links are internal links that say Home Page. If we simply removed those internal links the site would rank higher but a much better solution would be to change them into text links that say Hats.
• The #5 page would score better with a directory listing like Yahoo or ODP.
• By far, the #4 page is succeeding most with the least amount of effort.Why? …because it is doing everything right.
  1. It has a directory listing which is very good.
  2. In addition, our analysis (using OptiLink) revealed that almost all of the incoming text links contained the primary keyword – hats.
  3. AND a good percentage of these links originate on pages that are, themselves, about hats.
• The #3 page beats #4 mostly by brute force – the sheer quantity of incoming links.This proves that, in spite of zero occurrences on the page for the keyword hats combined with a lack of keyword in the title of the linking pages, quantity of links can win – especially when there is a fairly high percentage of keyword in the text links (in this case 64%) combined with a significantly relevant Yahoo listing.The downside of this type of scoring strategy is the difficulty of obtaining (or creating) 486 incoming links!This page would likely score at the top if they increased the percentage of keyword occurrences on the page itself combined with a linking strategy that focused on pages that have the primary keyword in their title.
• Finally, the #1 and #2 position (these are actually the same page) beat all of the others, again, mostly by brute force – shear link quantity.Contributing factors that help these pages are a respectably high percentage of keyword appearance on the page itself (in this case 16%) combined with a fairly high percentage of keyword occurrence in their incoming text links (in this case 56%).Of course, obtaining or creating 982 incoming links is a chore all by itself. Worth pointing out, however, is the fact that if this page were to get the other factors correct as well, i.e.,
  1. a higher percentage of keyword in linking title,
  2. closer to 100% keyword occurrence for incoming text-links, and
  3. a Yahoo and ODP listing,

  …then this site would be virtually unbeatable in the rankings!

Conclusion

So does Link Popularity work? Sure – but only if you overwhelm the competition with a brute force, sheer-quantity-of-links type strategy. As we’ve shown you, three of the pages in the top six for hats at Google have little more going for them other than link count – i.e., popularity.

Impressive, however, is the fact that the #4 page scored so well with so little link popularity – all because they focused correctly on building their link reputation. Based on what you’ve learned so far, which link infrastructure would you rather build: page #1 or page #4?

Personally, I’ll take #4 any day of the week because by the end of the following week I’ll have doubled the number of links. Through link solicitation, combined with building more good content on my internal pages that link to my target pages, in a very short time my page will be #1 anyway! …and with a LOT less effort.

More Info: OptiLink Link Reputation Analyzer http://www.windrosesoftware.com
Best regards, Leslie Rohde – Windrose Software Creator of OptiLink

�