 |
I have recently started tracking web traffic, moving beyond simple page hit counters. In the process, I have learned a bit about web rings. A few ringmasters e-mail ring statistics to their ring members. I recently learned that these statistics are available to the general public at http://stats.webring.org/. Thus, I can take a peek at what is going on in all my rings, or any ring. As I may do this from time to time, I set up the following table.
The left side hot points will bring one to the ring's home page. The stats hot points gives the 25 most active sites, sites that send traffic to the ring, and the 25 sites that receive the most hits from the ring. The activity hot points show the amount of activity generated by every site on the ring. The hits point tells how many hits each site on the ring receives. For laughs, I've included the usual ring controls as well. If you want to search on how well my site is doing, or your own, you may wish to have the site id number for the site, provided for my sites in the left column.
In early October 1999, I took a snapshot of the performance of most of my rings over a two month interval. I don't intend to update the table. I also don't intend to say there are good rings and bad rings. There are just different rings. Some rings are about subjects which attract the attention and passion of fans. These rings generate lots of traffic. Other rings seem to exist for other purposes. This is not an exhaustive list of ring types and flavors, either. These just happen to be the rings that seemed to belong with the things I write. Regardless, the numbers, sorted by the number of sites in a ring...
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Graphing out the data, one can ask a few questions. Do big rings or little rings generate more traffic? The following X-Y scatter chart plots the average amount of traffic generated per site against number of sites in the ring.
 |
|
|
There seems to be a broad trend towards larger rings generating more traffic. Most of the rings are in the lower left hand corner of the chart. Within that clump, the bigger rings tend towards more traffic. Outside the lower left corner, things get weird. The major lesson of the above chart is that certain subjects attract the attention of large numbers of fanatic followers. Ships, planes and Audrey Hepburn might do as examples. If you want traffic, deal with subjects that others care about.
On the down side are the Ring of Words and the WebRing Addicts. The Ring of Words is a serious and well managed ring. Their problem isn't with ring integrity. I popped "random" thirty times running and found the Ring of Words control box easily every time. They just have a broad subject matter. They accept any site with good writing. It is a writer's ring, not a ring focused for a specific subject that people care about. This is too bad, as there are some good sites on the Ring of Words. It is an example of a good ring with quality work that is focused towards the site owner's purposes, not towards a subject matter that attracts fans.
I suspect the WebRing Addicts WebRing was likely never intended to gather lots of traffic. I joined after seeing the ring control box at another site, laughing, and admitting to myself I belonged. Some people collect web rings like others collect baseball cards or coins. Do you know how many Disney rings there are out there? Given the high number of rings and limited number of fans, what might this imply about the traffic on each ring?
 |
|
|
My next question was whether people used next, index or random more to move from site to site. The answer is index, usually. The web is an interactive medium. People like to have control. If one is looking at a list of sites with descriptions of the site, one has a lot more of an idea where one is going than by hitting "next" or "random" blindly.
This is slightly more true of small rings than big rings. I see an upward trend in the blue line, above, while the red and green lines seem fairly flat. I'm guessing that in a small ring one can see the entire table of contents on one or two pages, so looking at a list is easier than with a big ring. The above inspires me to pay a little more attention to the title and description of my sites. A good number of ring users are looking at the names and descriptions, and going to the sites that catch their eye.
Still, there are lots of folks out there who use the next and random keys as well. Note that things change from ring to ring. I won't speculate on why Fan Fiction people read the list of sites, while the Friends of the Dragon hit next. I do wonder if I made a mistake jotting down the NavShips data, or whether they navigate with a sextant rather then next, index or random. I didn't record some categories, such as previous, next 5 and previous 5, but for most rings these options are seldom used.
 |
|
What percentage of sites get at least a hit a month? |
Finally, is one apt to get one's share of the ring's traffic? If one joins a big ring, and ends up on page 48 of the index, and the index is never rearranged, one might never get a hit from an index page. If the site immediately ahead of one has no ring code, one will never get traffic from a next click. If one is on page 48 of the index, and the guy before you has no ring code, all one can do is hope for a few random hits. A lot of people on a lot of rings get zero traffic from the rings.
Some rings are better than others at sharing traffic. My small sites are all over the place, from 15 to 90 percent of the sites get at least a hit a month. The mid sized rings are all pretty good. There is a huge difference between the two huge rings with over 1000 members. Most of this, I suspect, is subject matter. Airplanes draw more loyal fans than little known writers. If there are more hits to go around, fewer sites are left entirely without traffic.
I'm a programmer. I do C and Unix scripts for a living. I thus have this compulsive desire to see intact data structures. A web ring is a data structure. If you don't make sure your data structures are perfect, the program crashes. Thus, I'm perhaps a bit obsessive about making it easy for a visitor to my site to find the ring code so that he might continue his journey. Some of my rules of thumb...
Of course I have a large, complex site. If one's site deals in only one subject matter, and all one's rings are related to that subject matter, life can be much simpler. I might have someone come in on the Aero ring, jump to my section on poetry to read High Flight, then needs to find the Aero ring code again. Smaller sites don't have this problem. Some middle sized sites do, but the web spinner isn't aware of it.
If one wants traffic, one shouldn't ignore the surfers who prefer to use next5 or index. I just changed the name my Bryce 3D ring pages from "Image Gallery" to "Mirror Dance," the name of the first picture in the gallery. Why? If you want to get traffic from users who prefer the index method to the next method, one has to catch their attention. A little thought might go into what sort of people surf a given ring. What title and description is most apt to gather a little attention?
There are an awful lot of rings out there. When making a choice which ring to join, it might be a good idea to look at the statistics. Is there a small set of sites that get all the action? Did the bulk of the sites get 0 hits in the last eight weeks? If so, if your objective is to attract traffic, perhaps you should look at other rings that show better stats. If you want traffic stats for your own ring, or a ring you are considering joining, click any of the hits hot points in the above table. The URL will include the phrase "ring=xxx" where xxx is the ID tag for the ring. Replace xxx with the ring ID tag for the ring you wish to inspect.
There are other reasons to join a ring. Sometimes I write a web chapter that just belongs in a given ring, whether that ring is generating traffic or not. Dragon Friend belongs in The Friends of the Dragon Ring. The Right to Keep and Bear Arms belongs in the Right to Keep and Bear Arms ring. My android character Joy Mudd belongs in the Mary Sue Society. Some of these rings are healthy and bring in traffic, while others might remain small low traffic rings indefinitely. No apologies. If a site is a square hit for what I'm doing, I may join regardless of the anticipated traffic. Even if Mary Sue brings very little traffic to Joy Mudd, Joy belongs in the Mary Sue Society.
Some people just collect web rings. The more rings one belongs to, the better. The more sites in one's ring, the better. The object seems to be to die with the most toys. Thus, the ring statistics are a useful tool in selecting a ring. Is the ring a status symbol, or will it generate traffic? If your goal is to get hits, you might want to know.
Extreme Tracking also traces which search engines bring users to one's site. I didn't mention search engines in the above essay as I got not hits from search engines.
Now engines are wonderful things. I also know many of the automated search spiders have found my sites. I can search on "Joy Mudd" or "Glenn Blacow" on many search services, and sure enough, there I am. The problem is, how many people search on "Joy Mudd" or "Glenn Blacow?" I also have had two sites listed in Yahoo for years, Flight by Committee and An Android's Tale. Alas, they are buried in fairly obscure folders. How many people visit "Aviation Travelogues" or "Star Trek, Role playing games, AOL?" I've been getting a steady trickle of visitors to these sites from the Aero and Fan Fiction rings, but had seen no sign of search service hits. (Update, I recently got a hit from an Alta Vista search for gun control arguments.)
But, while I was chasing traffic, I decided I'd submit two more of my chapters to Yahoo. I Won't Wear It was suggested for "AD&D humor." Sickles' Hole was nominated for "Civil War, Battles and Campaigns, Gettysburg." Several days later, in looking at my statistics, I got a pleasant surprise. A search engine hit! Yahoo had accepted Sickles' Hole.
I've been used to getting zero to three hits a day from my more active web rings. This is a slow steady trickle. Over a week or a month, this builds up. The first full day on Yahoo, Sickles Hole drew 52 hits, and things are holding steady.
This likely won't last. I have an interest in the Battle of Midway. Every once in a while I visit Yahoo's page on the Pacific front, and see if a new site has been added. After the Gettysburg fanatics who occasionally visit Yahoo have seen Sickles' Hole once, the hit rate will likely drop down considerably.
Still, I can confirm that if you get a site in the right category at Yahoo, it is still the 500 pound gorilla of search services.
 |
|
|
The above chart is based on Extreme Tracking's "unique visitors" statistics on a given page, the teaser page for the chapter. In the beginning, only a few of my chapters were tracked. I am adding more chapters as I move them to the Polyticks domain. These statistics track hits on a chapter, not hits from a given ring. Polyticks (Political, shown in red) is a member of three rings. Recognizing the Battle Flag (blue) is in no rings - slavery seems to be too sensitive a subject for the Civil War ring masters - but it is a subchapter of Polyticks. I have filtered out most of the hits due to my visiting my own pages, but a few of them are there early in the chart.
Can you see the subtle change when Sickle's Hole got the Yahoo entry? The first low day was a Saturday. Shortly after, there were three low days in a row. These came as the Yahoo people moved me from the "new" position at the top of the page to the permanent slot in alphabetical order. For a while I thought the "new" position was worth 30 hits a day, but the hit rate is climbing again. We shall see how things steady up in the long term.
There is also a subtle increase in traffic to my other sites with Yahoo hits to Sickles' Hole. Before, I was generally getting zero to four hits per day per chapter. Six or more hits per day is more common after Yahoo. This can be confirmed by the more detailed statistics. An good size share of the hits on my non Sickles Hole sites are referrals from Sickles Hole. Once Sickles' Hole brings em in, I do try to tempt people to visit other chapters. To some extent this works.
The jump in the yellow-brown art ring came as I joined the 3D ring. 3D lists new sites first on their index page. Until even newer sites join, a new site has the most preferred position, and gets an extra share of ring traffic. This seems to have settled down a bit. 3D has joined the rest of the rings.
The purple is An Android's tale. I've only a week of statistics. For a while An Android's Tale was up there with Sickles' Hole. What happened? One of the first hits I traced came from a web site I'd never heard of, gynoid.com. They are into female androids, anything to do with female androids, but the origins of the community are in the news group alt.sex.fetish.robot (ASFR). My androids have an element of their genre in them. The Joy Class androids are based in part on Star Trek's "I Mudd" episode, where an over sexed male programmed and built a planet full of female androids exclusively for his pleasure. Still, there is far more to Joy than that.
I posted a few messages to gynoid.com's board. One of the better known people in the gynoid community posted a kind review of my page. The result was a massive spike in traffic. I may visit again, occasionally, and post comments to the board if any seem appropriate. Still, I'm not a precise match for that community. While Joy is within their sphere of interest, she isn't at the center of their sphere of interest. Going out of my way to create message board links into my site seems improper.
While the bulk of my Sickles' Hole Yahoo hits have been from their Battle of Gettysburg folder, I've also had a number of search report hits. When registering a Yahoo site, the request that one does not repeat the same words as the folder. Their folder is Arts > Humanities > History > U.S. History > 19th Century > Military History > Civil War (1861-1865) > Battles and Campaigns >Battle of Gettysburg. I avoided repeating as follows. Sickles' Hole - strategic overview, centered on the July 2 advance. Features photos and maps. All of the above are key words that contribute to getting a hit on one's site. I lucked out in using the words "photos" and "maps." I'm getting a lot of hits from "Gettysburg Maps," "Civil War Photos" and the like. The simple one word search "hole" is also very common. "Hole" is a rare enough word in the Yahoo directory that it is effective in isolating Sickles' Hole. I've also gotten hits from some rather unexpected searches, having nothing to do with Gettysburg or the Civil War. Just for laughs, a few of my favorites follow.
Perhaps the reason I get no hits from any search service other than Yahoo is because most of them don't know I exist? I'm just trying out the Add Me service, which makes it easy for folks to register with lots of search engines. It takes only a few seconds per search engine. It takes quite a while by the time you finish with all the engines. I'll post any change in traffic here. :-)
The down side is that they require a link back add. The first choice was too animated for my taste. Fortunately, there was a moderately sedate option available. The following is also the only add on my sites. We'll see if Add Me delivers...
You have to be added to add?
 |
Update... I have had something an increase in traffic from search engines other than Yahoo since running Add Me. Yahoo's Gettysburg reference to Sickles' Hole still out draws all other engines on all of my other sites, combined. I get more hits from rings than from the other search engines. Still, it was worth spending a few minutes walking through Add Me.
On the traffic by chapter chart above, there is a small spike in black lines with orange square data points. This spike is my chapter Visions, which is attached to no web rings. Visions is my only non-original chapter, mostly classic poems. It has no new content provided by me, I own the rights to none of the poems, so I didn't attach it to any rings.
I've only a week of data on Visions, and I'm only collecting stats from one poem. The spike peaking at 12 hits in one day comes almost entirely from Alta Vista searches on "Do not go gentle into that good night." While there are dozens of copies of that poem on the web, for some unknown reason, my copy is listed first on Alta Vista. Thus, I get lots of Yahoo hits for Sickles Hole, quite a few Alta Vista hits for Visions, and a scattering of hits from other engines for other chapters.
I intend to put traces on the rest of the pages on that chapter, just for laughs. How many hits am I getting for "Four scores and seven years ago," "I have a dream," "No man is an island," "I must go down to the sea again," or "For everything there is a season?" I must also put more hooks into Visions to make it easier for surfers to find the rest of my pages.
If you do try Add Me, be patient. A lot of search engines only update once every other week or so. The increase in traffic from the lesser known search engines took a week to take hold in earnest.
|
|
Vista |
|
smart |
|
crawler |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
My top key words include Gettysburg, hole, maps, war, of, civil, battle, go, not, do, gentle, map, good, night, into, that, photos, the, air, and, combat, world, tag, maneuver, sickles, on, american, sky, laser, warriors, america, flight, overview, military, day, rings, 19th, three and sick. While Yahoo has a huge lead in the above chart, this is exagerated somewhat as Sickles' Hole is one of first sites I put counters on. Yahoo had several weeks head start before I ran Add Me and planted statistic gathering hooks in more of my sites. Cutting the Yahoo hits in half is reasonable, but it still doesn't much reduce the gap.
I've also stumbled onto dmoz.org, striving to be "the largest human-edited directory of the web." The approach is similar to Yahoo. They have editors volunteer examining pages, making sure sites deserve to be in the listing, and maintaining hierarchy. The result is a good browsing directory. It seems easier to get listed in Dmoz than Yahoo, but Dmoz is less known and less visited.
Rhythyms
There are some obvious trends in hit rates. None of these drive me to write or promote pages differently, but the numbers might be of interest none the less.
 |
The first pattern shows more hits in the early weekdays than during the weekends. I figure I have better things to do on weekends, and this might be true in general. The vertical axis indicates the number of hits over a four month period.
 |
The second pattern shows the United States still dominates the web, at least for english speaking hobby oriented pages. The hours are shown in Eastern US time. Peak use about 9 AM to 9 PM.
 |
The final pattern shows weekly numbers over late 1999 to early 2000. Weeks in a year are numbered 1 through 52. Christmas and New Years weeks are obviously low. Again, people likely have other things to do. As for the rest, I wonder if bad weather results in more web activity? As spring and summer come around, and people get outdoors more, will traffic sag?
