MSN LinkExchange Flimflam
By: The Designer | 06Mar1999
|
To best understand my story, I ought to fill you in on banner swap programs in general -- in case you aren't familiar with the particulars: There are two major categories of banner exchange programs. You place code on your pages that allow a banner from the pool to show up. With the non-profit kind, you get an even swap. Each time someone else's appears on one of your pages, the banner you designed and submitted to the pool will appear on someone else's page. Then there are banner swaps that make a profit. Your banner shows up a percentage of the times that a banner displays on your site. Most commonly, you get one exposure for every two times another banner displays on your pages. The extra banner display credits are sold to advertisers. You provide content which draws in surfers, and someone else gets exposure -- either because they "earned" the credit or paid for it. Quest of the Unquietmind was a member of the LinkExchange banner swap program since we first published here in April 1998. Our "sister" site, Seeking the Aircooledmind, had been a member since February 1997.
In the log, sometimes it lists the URL of the page the banner was on and sometimes they generally look something like this: http://leader.linkexchange.com/1/X783349/showiframe? This particular link would be from banner "/1/" on LE member 783349's page. Even though it doesn't list the URL of the page, it isn't too difficult to determine where that site is located. In your browser's location field, add the member number to the end of http://bookmark.linkexchange.com/bookmark? and press "Enter". This bookmarks the site with www.surfpoint.com -- the categorized directory of LE members -- and you'll be sent to Surfpoint's Bookmark Manager, where you can click to the page from there. Many LE members have more than one LE banner showing, but if you click around, you can determine which is "/1/" or "/28/" by holding your pointer over the banner and reading the status at the bottom of most browsers. I found the page where our banner had appeared -- that someone had clicked to get to our site. It was at the bottom of the page. That didn't seem right! You see, we weren't just LE members, we were "Deluxe" LE members. From a LinkExchange page:
"If you want to ensure your own banner GIF appears in the first screenful of other members' web page(s), join our Deluxe program. To participate, you must also place the LE HTML code above the fold (first 480 pixels) of your web page(s)." I wondered why our banner had appeared at the bottom of a page. I thought maybe the requirements had changed, so I wrote email to admin@linkexchange.com:
Is it no longer a requirement to insert LE code on pages "above the fold"? A few days later I got a friendly reply:
Date: Tue, 29 Dec 1998 12:31:05 -0800 The rules hadn't changed, so it must be an anomaly -- I assumed. That wasn't my only assumption. Our click-through ratio on our banners had been better too, but I assumed that the surfing population was growing harder to please. Or something. I kept designing new banners and adjusting our targeting, but it didn't seem to help. Oh yeah, I should explain about targeting. That's the really crucial factor in placing banners -- the ability to target the sites on which your banner appears. This is why some banner swap programs are associated with a web directory like Surfpoint. One of LinkExchange's pages states it this way:
"Our on-line poll indicated that the one feature our members want more than any other is targeting. Categorizing your site lays the groundwork for targeting. Targeting gives you more control over where your ads are placed as well as which ads will appear on your own site." and "The value of targeting is in reaching a qualified audience, one you choose. If you maintain the same click thru between non-targeted and targeted ads, the targeted audience should be more valuable. Increased interest, longer visits, and greater response are the primary reasons for targeted advertising, not click-thru ratio." All LE members can categorize two places where their web sites will appear in Surfpoint, and can target two Surfpoint categories -- the same, or different from their own. For example, if your site is about collecting Teletubbies, you may not want to target sites in the category Religion: Christianity: Paranoid Windbags -- although I'm sure people with sites listed there would appreciate your bright, animated "Satan is my buddy!" banner on their pages. If you aren't satisfied with your click-through ratio, your banner might suck. So design a better one. Make sure what it says corresponds with the sites you are targeting. That's how it is supposed to work! A few weeks ago, I decided that I really ought to do something about our click-through ratio, so I'd try another banner. I emailed LinkExchange and asked them to suspend our banner displays and let the credits build up. When I had the time, I was going to try to put together a real attention-getter. I finally got around to it about two weeks ago. If you were already looking at a page in the category Society & Issues: Politics and Debate or Social Issues, would you click on this?
 I submitted it and it was approved (they check banners for suitability), then wrote LE again, and asked them to start displaying our banner again. Our two targeted categories had about 1800 sites between them, according to Surfpoint -- but remember, we were also "deluxe". The banner was supposed to show up only on other deluxe member's sites and in Surfpoint, deluxe members are listed in bold print. I estimate that around a third of the sites are deluxe, which would be about 600 sites. Also bear in mind that the program isn't going to show only our banner. Other people target the same categories -- and banners for LinkExchange's various services appear at least a third of the time. By the time our new banner did begin to display, we had nearly 7000 credits built up. So, how long should it take to use up 7000 credits on whatever number of pages are represented by about 600 web sites? Our first hit through a LE banner that our server logs captured was at 15:51; the last was at 18:30 on the same day. During that three hours, our banner got about 6900 exposures, and was clicked on 37 times. Not only was our click-through ratio as bad as ever, the time it took to go through all those credits seemed unreal. Now (finally) I was suspicious. So I seriously studied those server logs. They didn't pick up all 37 hits, but I traced thirteen of the hits back to the source -- a time consuming process -- before I got tired of looking. Of the thirteen, eleven did not conform to the kind of sites I'm targeting: deluxe members in the targeted categories. Seven of them were from neither of our categories. (Two more didn't seem to have a category in Surfpoint -- probably new members.) Nine of the thirteen were not deluxe member pages, by virtue of a lack of boldfaced listing or the banner being at the bottom of the page or both. I checked recent server logs for our sister site (another LE Deluxe member). Of the six I traced back, four produced results similar to what I found with this site. I've got to wonder, how many hits would I have gotten from our nearly 7000 credits if LinkExchange had been playing by the rules -- and how long have they not played by the rules. (Probably since our click-through ratio started down and never came up again.)
"Although we will make a reasonable effort to provide a high standard of quality for our services, LinkExchange makes no guarantees of any kind regarding the dependability, accuracy, or timeliness of the services of LinkExchange." There isn't a "reasonable effort" any longer. Prominently displaying another banner in exchange for a good display of your own has disappeared. In their own documentation they explain how important targeting is -- in case someone doesn't understand the presumed benefit of using their service -- but targeting has become non-existent. What's worse, they haven't informed their membership -- or paying advertisers -- that they don't actually target any more. Two days ago, I removed the code for displaying LE banners off every page but one, from this site and Seeking the Aircooledmind. One banner kept our membership open long enough for me to send email to the administrators of the MSN LinkExchange and tell them I had discovered their dirty little secret. If they answer that email with some kind of explanation -- now that I've terminated our relationship and removed the last banners -- I'll publish it in this space.
So if you are looking for a banner swap program, I suggest you look elsewhere. We haven't tried anything else -- and probably won't at this point but there are a pretty good list of them at Mozilla's Open Directory Project. One thing the LinkExchange account was reasonably good for was seeing what pages are hit, and when. But if you have access to your own cgi-bin directory, I recommend AXS, a Perl script available free from Fluid Dynamics. It's easy to set up, fast, and will give you access to your own server logs.
(Geek interest only.) |
If you aren't a web designer, this may be of little interest to you. If you do have a web site(s), and think that a banner swap program is a good way to get visitors, there's something you ought to know about MSN LinkExchange. "MSN" is the Microsoft Network, which bought LinkExchange a few months ago and I'll bet that has a lot to do with what has happened to what used to be a reasonable service.
I noticed something a little odd last December. I was going through the server logs for this site, checking out pages that linked to us. Viewing server logs are a good way to find out who believes a web site is worthy enough to make a link to it, or to see which keywords in which search engines are sucking people in. I don't normally look up pages that linked here through a LE banner because they are incidental, but for whatever reason, I traced a link back that came through a LE banner.
I don't think the people at LinkExchange are "bad". Months ago, our click-through ratio was much better. Something changed. As I wrote at the beginning, I believe what changed was that a decent little company doing a perfectly good service was bought out by the Microsoft Network. As you may have read