Unique Vists to the Web Site

This graphic shows the unique visits to interfacebus.com in three different views, in the same graph. The largest of the three shows monthly visits, month after month.

The smaller insert also shows unique web visits, but with the years overlapped one another. The smaller of the inserts also shows visits but as a year to year comparison.

The variations from month to month indicate some fluctuation in the incoming traffic. However the site receives several thousand visits per day, so many of those dips and peaks represent just one day of traffic, and some may just be the site counter going off line.

Some of the months perform better or worse than other months only due to the number of days in the month or holidays occurring during the month. A work day brings in several thousand visits, while a week-end may only bring in 50% of that. A holiday only drops the visits by 75%, mainly due to 45% of the visits coming from other countries.

This posting allows the blog to serve the graphic instead of my server, but the web site does point to this page.

Click the graphic to see the larger view.

Although it's not that relevant any longer, the large dip or drop in visits during 2005 was due to changing the web counter mid year.
.

Year End Site Performance

Last year ended up with the highest number of visits to the web site so far. Google would say that the total number of visits is not the best metric to watch, but I would contend that any visit is a good visit. However Google likes Time on Site or Bounce Rate, but a bounce [one page view] does not mean the visitor does not come back later.

The graph shows the number of visits to the site. The number of visits indicates a visit and/or a returning visit from the same person, as opposed to unique visits, when a returning visitor is not counted. The graph compares month-to-month visits and year-to-year visits. Except for December, this year out performed every month in all the previous years.

Around mid-year I finished implementing the new Google search bar. They say it's better {?}, but the code is smaller so I changed out the code. In 2008 I changed out all the Google banners, for that new code too, which was also smaller.

Any time I update a page I look for white space I can remove to reduce page loading time. White space is a 'space' in the html code that severs no function. Normally I can reduce a page by 200 bytes for a 10k html file and 2k for a 50k page by deleting [not visible] white space. The down side is that I also add new content when I up-date a page, so in many cases removing the white space evens out with the new text.

So it's been seven months now of optimizing the web site to increase Google's Site Performance data. Now, 7 months later the site's performance is 2.8 seconds [average load time] which happens to be what the loading time was back in June. However the average 'average' is closer to 2.8 than before, which hovered into the 3 second range.

Two weeks ago I started trading out the Google Analytic tracking code, which they also say is faster. So I hope to see a speed improvement in a few more weeks, as I get more pages running the new code. Currently there are just over 300 pages that are running the new Analytic code, but the Site Performance data is weighted so heavy trafficked pages count more. I up date pages based on need, not traffic.

Any way I'm always adding new content, regardless of any SEO things done to the site.