The Future of WebSite
Ranking
By
Mel Strocen
Copyright © 2003
The recent shakeup in Google's search results, which set the SEO (search
engine optimization) community buzzing and saw tens of thousands of webmasters
watch their site ranking plummet, was in many ways inevitable. Almost all
SEO companies and most savvy webmasters had a fairly good handle on what
Google considered important. And since SEO, by definition, is the art of
manipulating website ranking (not always with the best interests of searchers
in mind), it was only a matter of time until Google decided to make some
changes.
If you've been asleep at the SEO switch, here are a few links to articles
and forums that have focused on the recent changes at Google:
Articles:
http://www.sitepronews.com
http://www.searchengineguide.com
http://www.searchenginejournal.com
http://www.accordmarketing.com
Forums:
http://www.webmasterworld.com
http://www.jimworld.com
http://www.searchguild.com
To date, most of the commentary has been predictable, ranging from the critical
and analytical to the speculative.
Here's a typical example from one of our SiteProNews readers:
"I'm not sure what has happened to Google's vaunted algorithm, but searches
are now returning unrelated junk results as early as the second page and
even first page listings are a random collection of internal pages (not index
pages) from minor players in my industry (mostly re-sellers) vaguely related
to my highly-focused keyword search queries."
So, what is Google trying to accomplish? As one author put it, Google has
a "democratic" vision of the Web. Unfortunately for Google and the other
major search engines, those with a grasp of SEO techniques were beginning
to tarnish that vision by stacking the search result deck in favor of their
websites.
Search Engine Optimization or Ranking Manipulation?
Author and search engine expert, Barry Lloyd commented as follows: "Google
has seen their search engine results manipulated by SEOs to a significant
extent over the past few years. Their reliance on PageRank to grade
the authority of pages has led to the wholesale trading and buying of links
with the primary purpose of influencing rankings on Google rather than for
natural linking reasons."
Given Google's dominance of search and how important ranking well in Google
is to millions of websites, attempts at rank manipulation shouldn't come
as a surprise to anyone. For many, achieving a high site ranking is more
important than the hard work it takes to legitmately earn a good ranking.
The Problem with Current Site Ranking Methods
There will always be those who are more interested in the end result than
on how they get there and site ranking that is based on site content (links,
keywords, etc.) and interpreted by ranking algorithms will always be subject
to manipulation. Why? Because, for now, crawlers and algorithms lack the
intelligence to make informed judgements on site quality.
A short while ago, author, Mike Banks Valentine published an article entitled
"SEO
Mercilessly Murdered by Copywriters!." The article rightly pointed out
SEO's focus on making text and page structure "crawler friendly." Other SEO
authors have written at great length about the need for "text, text, text"
in page body content as well as in Meta, Heading, ALT, and Link tags. They
are all correct and yet they are all missing (or ignoring) the point which
is that the "tail is wagging the dog." Search engines are determining what
is relevant, not the people using those engines. Searchers are relegated
to the role of engine critics and webmasters to being students of SEO.
SEO manipulation will continue and thrive as long as search engines base
their algorithms on page and link analysis. The rules may change, but the
game will remain the same.
Therein lies the problem with all current search engine ranking algorithms.
SEO's will always attempt to position their sites at the top of search engine
results whether their sites deserve to be there or not, and search engines
will continue to tweak their algorithms in an attempt to eliminate SEO loopholes.
If there is a solution to this ongoing battle of vested interests, it won't
come from improving page content analysis.
Incorporating User Popularity
into Ranking Algorithms
The future of quality search results lies in harnessing the opinions of the
Internet masses - in other words, by tying search results and site ranking
to User Popularity. Google's "democratic" vision of the Web will never be
achieved by manipulating algorithm criteria based on content. It will only
be achieved by factoring in what is important to people, and people will
always remain the best judge of what that is. The true challenge for search
engines in the future is how to incorporate web searcher input and preferences
into their ranking algorithms.
Website ranking based on user popularity - the measurement of searcher visits
to a site, pages viewed, time viewed, etc. - will be far less subject to
manipulation and will ensure a more satisfying search experience. Why? Because
web sites that receive the kiss of approval from 10,000, 100,000 or a million
plus surfers a month are unlikely to disappoint new visitors. Although some
websites might achieve temporary spikes in popularity through link exchanges,
inflated or false claims, email marketing, pyramid schemes, etc., these spikes
would be almost impossible to sustain over the long-term. As Lincoln said
"You can fool some of the people all the time. You can fool all the people
some of the time. But you can't fool all the people all the time." Any effective
ranking system based on surfer input will inevitably be superior to current
systems.
To date, none of the major search engines have shown a serious interest in
incorporating user popularity into their ranking algorithms. As of this writing,
ExactSeek is
the only search engine that has implemented a site ranking algorithm based
on user popularity.
Resistance to change, however, is not the only reason user data hasn't made
its way into ranking algorithms. ExactSeek's new ranking algorithm was made
possible only as a result of its partner arrangement with Alexa Internet,
one of the oldest and largest aggregator's of user data on the Web. Alexa
has been collecting user data through its toolbar (downloaded over 10 million
times) since 1997 and is currently the only web entity with a large enough
user base to measure site popularity and evaluate user preferences in a
meaningful way.
The Challenges Facing User Popularity Based Ranking
1. The Collection Of User Data:
In order for web user data to play a significant role in search results and
site ranking, it would need to be gathered in sufficient volume and detail
to accurately reflect web user interests and choices. The surfing preferences
of a few million toolbar users would be meaningless when applied to a search
engine database of billions of web pages. Even Alexa, with its huge store
of user data, is only able to rank 3 to 4 million websites with any degree
of accuracy.
2. Privacy:
The collection of user data obviously has privacy implications. Privacy concerns
have become more of an issue in recent years and could hinder any attempt
to collect user data on a large scale. The surfing public would need to cooperate
in such an endeavor and be persuaded of the benefits.
3. Interest:
Web search continues to grow in popularity with more than 80% of Internet
users relying on search engines to find what they need. However, with the
exception of site owners who have a vested interest in site ranking, most
web searchers have not expressed any serious dissatisfaction with the overall
quality of search results delivered by the major engines. Harnessing the
cooperation and active participation of this latter and much larger group
would be difficult, if not impossible.
The future of web search and website ranking belongs in the hands of all
Internet users, but whether it ends up there depends on how willing they
are to participate in that future.
About the Author:
Mel Strocen
is CEO of the Jayde Online Network of websites. The Jayde network currently
consists of 12 websites, including
ExactSeek.com
and
SiteProNews.com
|
|