Jul < Ricci Green < Site Stats || search | sitemap | help gazette | theater | bistro

|

How many? other pages Parkside Plaza traffic report | visitors' domains

According to the folks who make Web Trends, whose software I use to analyze the server logs:

A request is what people usually mean by "hits". They are "all the successful hits including HTML pages, pictures, forms, scripts and files downloaded."

The page views are "the number of pages viewed and not including the supporting graphic files within the page."

A visitor session is a sequence of hits by a single visitor. Visitor sessions are considered over when no requests have been made for thirty minutes.

Unique visitors are a count of unique IP (Internet Protocol) numbers. Please do not associate visitors with people. Examples:

1) A visitor can be a spider or robot sent from a search engine.

2) Three students all work at the same company but only one has Internet access. They separately sit at the one computer and request pages from work the same afternoon. One or the other of them is at the computer all afternoon. They are all coming from the same IP number, so they will count here as one visitor and one long session.

3) Mary visits once each from work, home, and class all on the same day. That will count here as three unique visitors and three visitor sessions.

Clearly the raw numbers in the table mean less than the trends in the graphs.

Note the separate traffic report for the students' Parkside Plaza webs.

Requests

Visitor Sessions

Notes

The stats through June 1999 show only toLearn.net. The stats from July 1999 to June 2001 include RicciStreet.net as well as ClearLightStudio.com. The stats after June 2001 omit Clear Light Studio and include only toLearn.net and RicciStreet.net.

The toLearn.net server logs stay at a steady ten to fifteen megabytes per month all year. The Ricci Street traffic drops dramatically from mid-June to mid-August.

On the two graphs above, the curves are not the same shape. When school usage dropped off in summer 1999, traffic didn't in terms of visitors. These non-student visitors click away sooner but seem to return more often.

Month by month

Compared to August 1999, the number of requests more than triples in September 1999. However, the number of visitor sessions doesn't get close to doubling. A relatively few students stay longer.

Compared to September, the number of requests rises by about 13% in October, most of which is a fivefold rise in traffic on Clear Light Studio. However, the number of visitor sessions rises by half. More students not staying as long.

Compared to October, the number of requests rises by 40% in November. However, the number of visitor sessions barely rises. Clear Light Studio's numbers stayed steady, so the number of students tops out but they stay quite a bit longer.

Compared to November, all the numbers drop in December, reflecting the student vacation time. Requests dropped more than visitor sessions, so the visits didn't last as long and asked for more cached pages.

Compared to December, all the numbers rise in January 2000 to almost November's levels, reflecting the time off between mods.

Compared to January, all the numbers rise to new heights in February. Increased student participation, especially at the Bistro, accounts for most of the increase. Where the ratio of visits to visitors had fluctuated between two to one and three to one, it rose to almost four to one (19:5), indicating shorter visits to check the Bistro messages.

Compared to February, requests drop in March. However, visitor sessions is at an all-time high. More short visits indicate that the users' machines and their ISPs are getting a lot of cached pages.

Compared to March, the curve stays the same for both requests and visitor sessions in April and May. Consistent user behaviors and automatic caching are having consistent effects.

Compared to March, April, and May, requests don't drop as much as sessions in June. Until I get year-by-year comparisons, I won't be able to confirm my hunch that students keep checking over the summer, but they don't stay as long or look for new pages.

This hunch is reinforced by the number of unique visitors maintained in the far right hand column of the table above. Sessions have fallen to 65% of its March high. But unique visitors are still at about 80% of their April high. The same visitors are still coming, just not as often.

Compared to the spring, why did the numbers stay so high in July? Either lots of non-students request pages from the other two domain names on the server: toLearn and Clearlight Studio. Or lots of students are still using Ricci Street over the summer. As best I can estimate, Ricci Street got around a thousand page views during July from students. Next question: is that a lot or a little?

Compared to June and July, August is back up to the levels of March, April, and May. Almost all that traffic came after school began on August 23.

Compared to the previous year, September jumped from August to all time highs, exceeding February 2000. The increases came from more Bistro traffic and from Parkside Plaza, which wasn't active a year ago.

Compared to the previous year, October rose about the same. The increase is most noticeable at Parkside Plaza, for which you can now read a separate Parkside Plaza Traffic Report.

Compared to the previous year and to the previous month, November is down. I currently have one section of twenty students where last year and last month I had two sections of more than thirty.

Compared to November, December is back up, but not as far as October in either hits or sessions. The number of sessions was almost as high as October. The twenty students are more active than last year's thirty.

Compared to December, January 2001 is up again. The number of hits is almost an all-time high. The number of visitor sessions is an all-time high. More faculty members are using Ricci Street, especially the Bistro.

Compared to January, February 2001 is the highest so far. The number of visitor sessions is up only a little, but the number of hits is way up. People are visiting a little more often, and they're looking at more while they're here.

Compared to February, March 2001 traffic was down. The March numbers in these tables and graphs are estimates; one of the weekly log files got corrupted, probably by something I did, but it's too late to fix. While I know how large it was, I don't know the details and have thus posted March numbers that are eighty percent of February's and then rounded off. I suspect that the drop-off was due to Spring Break and the missed class sessions.

Compared to March, April 2001 traffic is back up close to February's records. More people are viewing fewer pages. More students are becoming experienced users. They are finding their way more efficiently and getting recently visited pages from the cache on their PC's.

Compared to April, May 2001 traffic shows a similar trend. About the same number of people visited fewer pages. The full report shows that Bistro traffic was down.