You can find potential sites by going to the search engines individually, but this takes a lot of time. Instead, go to http//FerretSoft.Com and download a copy of Web Ferret. It will search 9 search engines simultaneously and generate 1000 hits in about 10 minutes, if it finds that many.

It's free, and it has been one of the most useful tools on my desktop for nearly four years. If you want to get rid of the banner ads, register the program; it's only $25. One of the best bargains of which I'm aware.

Be sure you have Eliminate Duplicates selected under the Advanced option. Even so, you will get a bunch of URLs that point to the same site, but many will be eliminated. Next set an upper limit for your search. I use 999 for each search engine. Finally enter a search term such as people might use to find your site or one similar to it, and let the program run. Let it run in the background while you do something else.

When the run completes or you click Stop, save the file in .HTML format. Repeat the above with another keyword or phrase, saving the file to a different name. Repeat the process 3-7 times. However, a million listings is impractical. If necessary, use more specific keyword phrases to decrease the number of hits.

Next, collect all files to a single file in an HTML editor. Start at the top of it and copy the first base URL found into the search function. By base URL, I mean given http//home.com/index.htm, search on home.com only. This will find additional references to pages on this site such as http//home.com/about.htm.

Then work clear to the end of the file deleting all further occurrences of this URL. This takes time if the list is lengthy, but it assures no duplicates. I suspect that with so many sites into site support of some kind, it may take me several days to build a list appropriate for my site. But in the end it saves a much larger chunk of time, wasted in visiting sites you have already seen.

Given a list of sites that seem appropriate candidates, I load the page into my browser and one by one click off to visit each. Here's what I look for.

• A links page or the equivalent. If they do not have one or it is not easy to find, a link from this site will be of little value.
• Is the site a possible candidate? My personal criteria is that it be a site of interest to my visitors. And yes, I do include sites that directly compete with me.
• The size or importance of the site is of little interest. Lots of small sites grow. I've found that smaller sites often generate more hits than larger ones on which the links page is hard to find.
• If I feel the site will work, I then go looking for a contact name. If I do not find one, I use the contact info and ask a reasonable question in hopes of getting a name. There is simply no better way to begin a pitch for a link than with a name; "webmaster" doesn't make it.
• I poke about the site a bit, and always at least get off the home page, looking for something I like, for I want to include a positive comment in my request for a link.
• Since I prefer to use icons with annotated links on my site, I also look for an ad sized button. (If I don't find one, I'll ask for a URL when I email my request.)

Actually, all of the above kind of happens as I wander about. Given no links page, or one on which links are not annotated, I leave within a minute or two. And usually by then I have already decided whether or not the site is a candidate. When I have noted the information I need, I move on to the next site.

You bet, particularly if you have a long list. It is certainly not something you are likely to finish in a day. The trick is to set aside a bit of time each week and go through the process for say 10 or 20 sites. Then get on with things that must be done.

Look at it this way. If you identify 10 sites a week, and stick with it, you will have contacted 500 sites in a year. And if you contact each site correctly, you are likely to have something close to 400 links back to your site.

Before contacting a site, set up a link to it on your site. Some will argue you can generate links simply by asking. I have not had much luck with this approach. I have far greater success when I create the link to the candidate site before making contact. It allows me to open with a very strong note, for I have already done something that will benefit the site.

Note that if I'm working in chunks of 10 sites at a sitting, I suspend site checking given 10 candidates. I then create 10 links and upload the page, making note of where on the page the link can be found. Then I send a message.

I craft the message as carefully as I would a press release. All I need to do before sending it is ...

• Add a name after "Hi" in the first line.
• Work in mention of what I found on the site that I liked, which may mean some editing to make it fit.
• Where to find the link I have added.

People are busy. Period. And they are not really interested in your pet project unless there is something in it for them. My message will be as brief as possible and will mention *only* the benefits *to* the person I am contacting. I point out that I have a link that hopefully is generating more hits right now.

I wrap with an invitation to link to my site, but again stress the benefits in doing so. I make no mention of the benefits to me, for they are obvious. I attach an ad-sized icon that draws well, include a brief description of my site as might be appropriate in an annotated listing and the HTML for the link.

If I've written a good message, I generally get about 80% positive response, often enthusiastic with "thank you" used liberally. While you may consider following up on those who do not respond, I do not. In the past, those I checked with were not interested, so I assume no interest if there is no reply.

I have read of several ways to track traffic to assure that you do not have more visitors leaving your site than are coming to it. It is easy enough to do. For example, Count Link is a simple free CGI script from http//datatrendsoftware.com. It will track the number of times a link is clicked. But having tried a similar script, I don't feel the time is justified. I don't believe it is important that Joe gets 10 hits from my site while I get only 5 from him. My experience suggests it averages out over time.

About twice a year, though, I do check my log files. When I find zero hits from a site I have linked to, I quietly delete that link. It's a workable plan that requires minimal time, yet it assures me I am not carrying freeloaders.