Because Google is not the only search engine... [Archive]

View Full Version : Because Google is not the only search engine...

minstrel

11-12-2003, 10:22 AM

... and may not even be #1 forever, it is worth putting some effort into how you rank on the others.

Here, from a recent newsletter, is some information on optimizing for ExactSeek, using Alexa:

How ExactSeek's New Ranking Algorithm Works

The main factors used to determine website ranking are page content relevancy and site traffic data provided by Alexa Internet, with the emphasis being on content relevancy.

Title, Meta Description and Meta Keywords tags will continue to play a role in site ranking, but not in the way they they did previously. The information from the Title and Meta Description tags (if found by the ExactSeek crawler) will be shown in the search result listings, and the content of all 3 tags (if found) will be compared to page content as a check of overall relevancy. In cases where sites have more or less equal content relevancy for specific keyword searches, Alexa popularity data will determine which sites receive higher ranking in ExactSeek's search results.

How You Can Improve Your Site Ranking

The good news is that almost any traffic building strategy will work to improve your site's ExactSeek ranking, including:

-Link exchanges
-Any type of advertising
-PPC campaigns, and
-Search engine optimization techniques.

Why is the above true? Because an increase in people traffic to your website will result in a corresponding jump in your site's Alexa ranking which in turn will be factored into your ExactSeek site ranking.

Steven Glover

11-12-2003, 01:17 PM

I am a tad concerned about any SE that basis it's result in whole or in part upon a sites alexa ranking. I have taken the following information directly from Alexa in regards to the way it measures traffic.

Some Important Disclaimers
The traffic data are based on the set of Alexa users, which may not be a representative sample of the global Internet population. Known biases include (but are likely not limited to) the following:

Our users are disproportionately likely to visit alexa.com, amazon.com and archive.org, and traffic to these sites may be substantially overcounted.

The Alexa Toolbar works only with the Internet Explorer browser. Sites frequented mainly by users of other browsers will be undercounted. For example, the AOL/Netscape browser is not supported, which means that Alexa collects little data from AOL users, and our traffic to aol.com is likely lower than it would be for a more representative sample.

The Alexa Toolbar works only on Windows operating systems. Although a large majority of the Internet population currently use Windows, traffic to any sites which are disproportionately visited by users of other operating systems will be undercounted.

The rate of adoption of Alexa software in different parts of the world may vary widely due to advertising locality, language, and other geographic and cultural factors. For example, to some extent the prominence of Korean sites among our top-ranked sites reflects known high rates of general Internet usage in South Korea, but there may also be a dispropotionate number of Korean Alexa users.

In some cases traffic data may also be adversely affected by our "site" definitions. With tens of millions of hosts on the Internet, our automated procedures for determining which hosts are serving the "same" content may be incorrect and/or out-of-date. Similarly, the determinations of domains and home pages may not always be accurate. When these determinations change (as they do periodically), there may be sudden artificial changes in the Alexa traffic rankings for some sites as a consequence.

The Alexa Toolbar turns itself off on secure pages (https:). Sites with more secure page views will be under-represented in the Alexa traffic data.

rlrouse

11-12-2003, 05:43 PM

While it's true that Alexa rankings aren't "accurate" (and they never will be given the way the data are collected), they are a pretty reliable indicator of a site's traffic level relative to other sites in the same category or those that target the same search terms.

Google PageRank is far from perfect, but it's used every day by droves of webmasters to help them decide which sites to link to and which to avoid.

The Google SERPS for any given keyword aren't perfect either. Just look at the number of SPAMed sites and non-relevant sites that keep cropping up. But we still use them all the while we're complaining because we're happy with the quality of the SERPS overall.

My point is that it's virtually impossible to have perfect search results based on any ranking criteria.

IMO, search engine rankings based on Alexa data are probably just as "accurate" and "relevant" as any other. At least you're comparing apples to apples because you're comparing sites against other sites in the same "market".

Any site can SPAM it's way to the top of the SERPS for an unrelevant search term (at least for a while). But Alexa rankings reflect real visitors for the most part. The accuracy of the Alexa rankings themselves aren't really an issue.

The relative Alexa rankings of similar sites are an issue, and they can be pretty reliable when used in this context (for comparison purposes).

Of course only time will tell if my theory holds water...

minstrel

11-12-2003, 09:04 PM

I'm not crazy about ranking sites based on the Alexa toolbar either - it's a good example of sampling bias (the opposite of a random or representational sampling procedure), one of the first principles you learn about in any course on statistics or research methods.

However, we have to live with what's out there and apparently Alexa-based ranking is one of the things that's out there...

minstrel

11-12-2003, 09:14 PM

While it's true that Alexa rankings aren't "accurate" (and they never will be given the way the data are collected), they are a pretty reliable indicator of a site's traffic level relative to other sites in the same category or those that target the same search terms.

(snip)

IMO, search engine rankings based on Alexa data are probably just as "accurate" and "relevant" as any other. At least you're comparing apples to apples because you're comparing sites against other sites in the same "market".

But see my post above: Alexa rankings are "accurate", even within your stated limitations on defining accuracy, only for the population represented by people willing to install Alexa's toolbar on their browser, and that may be very different from anything we would be happy to describe as the "normal population" of internet users. You're right that no system is perfect but some are more imperfect than others. Would you be happy if your choice of television shows or movies was based on the viewing preferences of only those members of the population willing to install, say, a text-captioning device or some other non-standard accessory on their television receivers?

rlrouse

11-12-2003, 09:16 PM

minstrel:

I see your point, and it's one that I hadn't considered.

Steven Glover

11-13-2003, 08:36 AM

Consider this as well. We used to use Alexa as a measuring tool for ourselves. Ever computer here had the bar installed. Our homepage was probably loaded at least once per minute on every computer here. Our numbers kept getting better on Alexa. So one day we decided that we wanted to see what would happen if we took Alexa off all teh computers. SO I went thru unistalled Alexa and ran Ad-Aware to make sure it was gone. We left alexa off for two weeks and our Daily Page Count was nearly twice what it had been with the bar installed on our machines. We stuck the bars back on and it was an immediate reversal.

My point is it would be very easy to setup a computer or two with the task of loading the same page over and over again!

rocky1

11-13-2003, 12:32 PM

Getting away from the technicalities involved in this matter momentarily, one might also want to consider this from the logical standpoint that if a site is ranked based primarly upon it's popularity, how is a new site supposed to attain ranking on that Search Engine? Theoretically if it's new it's not popular; if it's not popular it's buried at the bottom of the list; if it's buried at the bottom of the list it isn't going to get much traffic, and thus it isn't going to get any more popular!

Unless you're on the top of the pile when they implement, you probably aren't going to get there without targeting the other engines, and doing some heavy marketing on your own.

In reality, if we all treat ExactSeek with the same respect they're showing us, they aren't going to be a contender for the #1 spot, and thus we don't have to worry about them, Alexa, or their algorithm!

Rocky

Steven Glover

11-13-2003, 12:41 PM

I hadnt thought of that point Rocky. You are correct!

computers

11-13-2003, 03:51 PM

A couple of weeks ago my site was #37 on Google, now it is at 107 and not moving up, very frustrating. I know that Alexa and ExactSeek are 'different', but I'm not listed on Alexa at all, but am in #6 on ExactSeek. Now my average monthly count is averaging about 2500/mth. Not too good, unfortunately ExactSeek is the best hope I have right now, I get most of my hits through them. Without the funds to do marketing, how is one supposed to get up there? I've checked and rechecked, even had my site checked, and it always comes out great, but I'm just not getting the high listing like I do with ExactSeek.

Conficio

11-13-2003, 05:56 PM

Thanks Rocky,
very good point! I'd like to expand on this.

What are we expecting from a good search engine? do we expect to be pampered with what is already popular? Does popularity equal quality?

Who thinks Hamburgers from McDonalds or Burger King are popular food, or Donuts from Krispy Kreme for that matter? Well they clearly are!

Who thinks these food are quality food? What quality do I expect? Well the incrediences should be fresh and the preparation should be hygenic. I believe they are, I can even watch the Donuts at Chrispy Kreme rolling down the coveyor belt.

Who thinks they are good quality in terms of healthy food? Well they provide me with some calories, but not a great diversity or balance for nutritional value. They all certainly lack vitamins and essential minerals, my body needs.

Who thinks they are a quality meal in terms of tingling my taste buds and providing a well roeunded experience, that I might enjoy and remember. Well they do, if consumed sporadically (fat and sugar was always a graving of mine), but I couldn't eat them every day or even more than once a day.

The point I try to make here is: Search engines should deliver quality answers to my query, not what other people think is hip or popular. This might be an interesting aspect of research or for comparison (such as how popular is my competition or who are the most popular three in a certain field?), but for quality results it does not really matter.

Also quality is not just a one dimensional measurement. As in the food example I can measure different dimensions of quality and not be able to compare one over the other (but I can weigh them according to a particular goal in mind).

Therefore I'd really stay away from any attempt to mix popularity in the ranking. And Alexa allows me to retrieve the info about popularity already. So I do not need ExactSeek to slant my search results. I'd rather like the apparent google approach, to put new sites on top of the heap and let them sink on their merrits (link popularity, content network, content actuality, subject focus). But the best of all worlds is diversity in search engines and algorythms. Because in different situations I find different SE do best. So lets keep discussing new approaches and lets spread the word about those that give us (and others) quality results = value.

My five cents
K<o>

mm99

11-14-2003, 05:21 PM

A couple of things...

I'm not sure how the new exactseek will pan out, but I do know a couple of things about the here and now.

To member "computers":

Right now, exseek just reads your tags (title, kws, desc). It's a dir as such, more than an se. I typed in your url and could not find you, even under just "customcomputerhaven", but when I broke it into 3 words, you came up 5 times. The way exseek finds sites is by desc and title and that is pretty much it. Not your url or content. Note the highlited words in your desc when you enter the 3 words. This is consistant with my past exp. For some reason, exseek does not weed out dupes. You come up 5 times under the 3 words.

The reason you are probably having probs w/ goog is because you code is very dirty. I fount this before your title tag:

<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns="http://www.w3.org/TR/REC-html40">

indicating to me, you used ms office to code your site.
Here's a random snip of code:

<v:f eqn="sum 21600 0 @3"/>
<v:f eqn="if @0 @3 0"/>
<v:f eqn="if @0 21600 @1"/>
<v:f eqn="if @0 0 @2"/>
<v:f eqn="if @0 @4 21600"/>
<v:f eqn="mid @5 @6"/>
<v:f eqn="mid @8 @5"/>
<v:f eqn="mid @7 @8"/>
<v:f eqn="mid @6 @7"/>
<v:f eqn="sum @6 0 @5"/>
</v:formulas>
<v:path textpathok="t" o:connecttype="custom" o:connectlocs="@9,0;@10,10800;@11,21600;@12,10800"
o:connectangles="270,180,90,0"/>
<v:textpath on="t" fitshape="t"/>
<v:handles>
<v:h position="#0,bottomRight" xrange="6629,14971"/>
</v:handles>
<o:lock v:ext="edit" text="t" shapetype="t"/>
</v:shapetype><v:shape id="_x0000_s1118" type="#_x0000_t136" alt="CCH" style='width:163.5pt;
height:74.25pt' fillcolor="#069" stroked="f" strokecolor="#369"
strokeweight="1.5pt">
<v:shadow on="t" color="#b2b2b2" opacity="52429f" offset="3pt"/>
<v:textpath style='font-family:"Verdana";font-weight:bold;v-text-kern:t'
trim="t" fitpath="t" string="CCH"/>
</v:shape><![endif]--><![if !vml]>index_files/image001.gif<![endif]> </td>
<td align="center" width="953" bgcolor="#E8EDF4">
<p align="center"><![if !vml]>index_files/image002.gif<![endif]></p>

I don't even know what much of this even means. If you recode it with something that will render better code, you'll do much better with googs.

The reason you do well with exseek is because it is just reading the head info it wants, goog will read the entire doc and your code is very bloated. Your html code size alone is 41761 bytes (that's a lot) which is almost half of the 104565 total size (also a lot).

Other matters:

I'm really wondering about something that maybe someone might comment on.

With Alexa, I looked up a site of mine located in brinkster.com. Since the site is a sub dir (not domain) it ranked me high on the scale. Looks like I'll do well with just about anything I have in the account, if the Alexa ranking is a factor.

I am speaking in the here and now and not what they are "going to do". This could be an interesting glitch. If I have a redirect "paulssite.freeredir.com" going to "brinkster.com/paulsstuff/index.htm" I can see the redir getting nowhere but the root, if submitted, ranking rather high. Comment?

peace...Paul