January 1997 Issue, copyright
1997, Canada Computer Paper Inc.
Weaving your own Web site: Part XVI
Publicizing your Web site: Part II
By Keith Schengili-Roberts
In December's "Weaving your own Web site" article we looked at
how you could publicize your Web site using traditional methods: contacting
other Web masters and setting up reciprocal links, using banner ads, and
the pros and cons of "Web awards." But these days, if somebody
is looking for a site such as yours, they'll probably start their hunt via
Yahoo, Alta Vista, Lycos, or any of the dozens of other popular search engines
available. How do you get your listing in a Web search engine, and how do
you make it appear prominent when somebody does a search on sites related
to your own?
Getting listed
Most search engines allow users to enter new uniform resource locators
(URLs) into their database. This process is usually pretty simple: click
on a "Submit URL" (or something similar) button within the search
engine and type in your URL. Many search engines will take things from there,
and send out a "Web crawler" program a few days later to check
out and automatically catalog your site. Other Web directories, like Yahoo,
ask you to write a brief description of your site, and what category (or
categories) you think it belongs in.
When you end the process, the Web search engine will typically check to
make sure that a Web site does exist at the URL specified (you did spell
everything right, didn't you?) and then the entry is added. In the case
of Yahoo, a person will check out your site, determine whether or not it
meets the particular search engine's criteria of being a useful site to
include, perhaps do some editorializing on your Web site summary, and then,
if your site passes the test, add it to its burgeoning database.
It is worthwhile to do some research before submitting a URL. If a search
engine uses category listings, check to see which category best fits your
Web site. If you find sites similar to your own in a category, you've found
the right place to register yours. Do not assume a broad category like "entertainment"
means the same thing for every search directory. Don't be shy to suggest
a new category if you can't find one to match yours; the chances of your
Web pages being truly unique diminish over time as the Web grows, so if
no other category or sub-category quite fits the topic of your Web page,
you may simply be the first to suggest the category.
While the process of adding URLs to individual search engines worked well
when there were only a few Web search engines available, these days there
are dozens of search engines available. While the majority of Web surfers
tend to stick with one of the more popular search engines (Yahoo, Lycos,
Alta Vista, Excite, Webcrawler, etc.), to draw in as many visitors as possible
to your Web site, you want as much coverage as possible. If you target all
of the major search engines, you can be sure that you will get some coverage.
There are services designed to partially automate the process of registering
your URL with many different search engines. One of the most popular of
these is Submit It! (http://www.submit-it.com/), a submission service that
enables you to register your URL simultaneously at several search engines.
You fill in a single, master form, select the search engines to post to,
and it takes care of the rest. Like many such services, it now offers professional
(i.e. pay-for-use) services for commercial sites, but it still offers a
free service that enables users to register their URL with as many as 16
search engines and directories at once.
As search engines proliferate, many are targeting themselves at niche audiences,
so now there are search engines being devoted exclusively to specific types
of Web sites, like music (Music Search, http://musicsearch.com/) science
fiction (Nebula Search, http://members.gnn.com/mochaman/nebsearch.html),
financial (Money$earch, http://www.moneysearch.com) and many more. Go to
a couple of search engines and type in the keywords "search engine"
and a keyword reflecting the topic covered by your Web site. If you find
a match, check out the search engine, and get your URL registered there.
HTML search engine tricks
Many of the search engines obtain URLs for their databases by sending
out programs known as "Web crawlers," that go out and search for
new URLs to add to the database. There is no human intervention-the Web
crawler finds new Web pages, retrieves them, searches them for keywords,
and then automatically adds them to the database. Knowing how these search
engines work can help you bring your Web page to greater prominence within
a Web search engine.
Since most Web crawlers end up searching for and storing pages by keywords
that appear in the page, it makes sense to ensure that a keyword or keywords
appear prominently within a Web page. At the very least, try to include
an important keyword as the title of your Web page, and within the <TITLE>
tags.
It has also been found that adding a description for the Web page using
the <META> tag helps several Web crawlers to properly identify and
catalog your Web page. This is not surprising, as one of the original reasons
for the creation of the <META> tag was cataloging purposes. Use the
NAME attribute of the <META> tag along with the CONTENT attribute
and a brief description, such as in the following example:
<HEAD>
<META NAME="description" CONTENT="Keith's Budgie Emporium">
...
</HEAD>
In general, it has been found that the more often a keyword appears within
a Web page, the more prominent a listing it will get within a search engine.
When a person searches for a common keyword term, most Web engines place
priority on Web pages that contain multiple instances of that keyword. All
things being equal, a Web page that uses the word "budgie" 10
times within a single page will appear higher in a user's search for budgie-related
Web sites than a Web page that only contains nine instances of the word.
This practice can be taken to extremes-overt repetition of keywords is known
as "search engine spamming." If you have ever seen a Web page
with a number of keywords either hidden from view in the header of a Web
page, or worse, visible within the body of a Web page, then you've run across
a case of it. This technique is frowned upon, not only because it creates
bulkier Web pages that take longer for the user to download, but because
Web crawler designers are getting smart, and are changing the characteristics
of their crawlers to compensate. While this technique may have led to a
top listing within a search engine in the past, spammed pages are often
passed over by Web crawlers-the designers have a vested interest in not
promoting this technique.
Too many visitors?
Keep in mind that there can be some downsides to creating a very popular
Web site. Take the case of a friend of mine who created a Web site that
became popular overnight. At the end of the month, he got a whopping bill
from his Internet service provider (ISP). It turns out that he was being
charged for something he had no direct control over-the total amount of
downloads from his Web site. His site made use of a lot of pictures and
text, so whenever anybody visited his site, they "downloaded"
all of the text and pictures to their systems. He was allotted 100 MB of
free downloads from visitors, but was charged per MB for everything above
that amount. Luckily he was able to work out something with his ISP regarding
his site, and he immediately reduced the graphic content of his Web site,
but it pays to check with your ISP to see if there are any hidden charges
when you create a Web site.
Keith Schengili-Roberts welcomes any comments, suggestions or HTML tips
and tricks you may have. You can e-mail him at: kschengi@interlog.com. His
book on HTML, The Advanced HTML Companion is available in bookstores.
Previous articles in the series can be found at The Computer Paper Web site:
http://tcp.ca/
Back to
Table of Contents...
Return to
TCP Home Page
TCP Online January 1997 Issue