PageViews performance

Yep I went out and looked at the performance of the new pages that have been generated so far this year. In most cases the page views are very low, maybe one or two a day or no views at all.

I opened three or four of the different pages and added a bit more text when possible. But a number of these pages were generated based on a graphic. So either the text is already embedded in the picture or the graphic doesn't really require any additional text. So I'm kind of stuck, the pages are un-fixable  there's really nothing wrong with them. Except for the fact that the page bring in zero traffic.

Now I know I need text on a page to bring in traffic from the search engines. It's standard Search Engine Optimization [SEO] stuff, day one. But these pages were generated around a graphic file, little text required. Maybe I should stop generating a new page just because I have a picture file I want to use. Still I'm not even getting hits from people already on the site [engineering], nor am I getting any traffic from people doing an image search.

The question is what to do? Some of these pages were generated 3 months ago and have only seen a dozen hits.
1. Well I added some additional text to a few of the pages.
2. I also commented on a few of the Blog pages that added them. I always comment on my own postings, to indicate updates or changes in the original posting. Of course the comment enhances the blog page, because more text is added. Remember a blog post is also a web page.
3. Then there is this blog posting, with links to the pages that need help.

Now I talked about this same issue last December [New Page Generation and Page Views], only how new pages did over the entire year. Normally I never care about a page until it's at least 3 months old, which some of these are. But these page views are so low there's just no getting around the fact that they most have some kind of issue.

Panel Mount LED. Holds a graphic of a few LEDs. Added 1/7/10, zero page rank
Capacitor Networks. Graphics and a bit of text. Added 1/15/10, zero page rank
Via Stubs in PCBs. Definition of a Via Stub. Added 1/17/10, zero page rank
Jumper Headers. Holds a graphic of some jumpers. Added 12/31/09, zero page rank
These pages have only received between a dozen and 2 dozen page views so far, subtract a few because of me. Oh and page rank doesn't mean any thing, but it wouldn't hurt if they get one soon. Of course there are more pages in the same boat, but why fill the blog post with a bunch of page links.

Pageviews still on the Increase

With the up-coming holiday I assumed that interfacebus would start seeing a decrease in site visits. However that has yet to occur, at least that I can tell. Yesterday was the second highest Tuesday on record, with 9,221 visits. Of course I'll see a decrease the day of the holiday, but because the web is world wide the decrease won't be total. In the early hours eastern Europe will be showing up, followed by South America and Canada, and then Western Europe ~ or something like that. As the chart shows the site just past 2 million visits to date for the year.

All the bad links were fixed from the site-wide check the other day, nothing but routine updates occurring now. There is a slow html enhancement being made to the Dictionary and Acronyms pages, but that will go on for weeks. In a few cases the page might visibly change but not by much.

I did notice that the main Transistor Derating page lost its page rank, not really sure what happened there. A few of those pages were also just updated [2N3441], but they still get almost no incoming visits. Guess everyone already knows how to derate a transistor, or I pick transistors nobody uses.

Graphic; Google Analytics / Map Overview / Countries/Territories 2009.

What is the Android Operating System

Android is a mobile operating system running on the Linux kernel. Android was started by Google, but it's open source now. Seems like it's primarily only on mobile phones right now but I bet that will change next year. Looks like there are a number of phones about to come out using or based on Android. Once the people making phones see a reduction of ~ $40 a phone because the OS is free, other mobile devices should start jumping on board. Why not, reduce your products cost by $40 or make another $40 in profit, plus it appears Windows mobile has a low user rating [or so I find on the internet].

Although the graph shows 2009 data, hits from Android started coming in last November, with 11 visits for the month. Visits have ramped up over the last year to 38 visits for last month [87% of which were new visitors], according to Google Analytics.

The server side counter [AWSTATS] does not report any hits from Android. However it might just be lumping Android visits under its "GNU Linux category [there are 7 other Linux based operating systems listed in the report].
Here are the different versions of Android coming in [2009]
Version 1.0 = 7%
Version 1.1 = 26%
Version 1.5 = 58%
Version 1.6 = 8%

Now it wasn't but a few years ago that search engines were telling web masters that they had to reformat their sites to allow them to work on mobile devices. Now, the un-formated sites seem to work just fine on those new devices, so I'm kind of glad I didn't spend three man years building another site just to find out later it wasn't needed.

Side topic; So far this month there are 138,782 unique visitors; higher than Jun, Jul and Aug.
By the end of the day there should be around 146,000 visitors, pushing it above Jan, Feb, Apr, and May.
Wednesday will pull in another 8,000 plus visitors. This month may bring in the highest amount of visitors this year.

Graph is produced from Google Analytics data.

Top Content and Page Views

I took a quick look at how the page views for the site were coming in.
The data shows the number of pages that get some amount of visits so far this year
Page views for 1/1/09 to 10/18/09

110,000 to 119,000: 1 page
100,000 to 109,000: 0 pages
90,000 to 99,000: 0 pages
80,000 to 89,000: 1 page
70,000 to 79,000: 0 pages
60,000 to 69,000: 2 pages
50,000 to 59,000: 1 page
40,000 to 49,000: 0 page
30,000 to 39,000: 4 pages
20,000 to 29,000: 10 pages
10,000 to 19,000: 50 pages
1,000 to 9,000: 458 pages
100 to 900: 869 pages
1 to 99: 350 pages [estimate]
So far there has been 1,401,430 unique visits to interfacebus.

err, ok I already know the under performing pages;
100 pages relating to How to Derate a Transistor.
100 pages relating to Manufacturers alphabetic Listing.
50 new pages, then just pages that only hold a large graphic.

Related Blog Posts.
Top Web Site Content  6/23/09
Top Content and Page Views  4/18/08

Are noon and midnight 12 a.m. or 12 p.m.?
The answer is that the terms 12 a.m. and 12 p.m. are wrong and should not be used.
      To illustrate this, consider that "a.m" and "p.m." are abbreviations for "ante meridiem" and "post meridiem." They mean "before noon" and "after noon," respectively. Noon is neither before or after noon; it is simply noon. Therefore, neither the "a.m." nor "p.m." designation is correct. On the other hand, midnight is both 12 hours before noon and 12 hours after noon. Therefore, either 12 a.m. or 12 p.m. could work as a designation for midnight, but both would be ambiguous as to the date intended.

Comparing Visits from Different Years

One of the reports from Google Analytics is an overview of visitors from any two dates that are of interest. You can also compare any two sets of date ranges. In this case the graphic to the left compares 1-1-08 to 7-22-08 vs. 1-1-09 to 7-22-09. The graph points [dots] are per week, but you can show data per day or per month. Click the image for a larger view of the data.

Notice that the graph lines appear almost identical. However the data indicates there has been a -0.56% increase visitors from this year to last, or around a decrease of 7000 visitors. That's less than one day worth of visitors, so maybe the site went down a few hours more this year than last.

Not much to worry about until I scroll down the report and review the data on a per page basis. That's when the data starts to look troubling. Here is a sampling of a few of the pages that are in decline, ignoring the few pages that showed an increase in visits.

CANbus -4.24% in visitors.
USB Interface -5.21% in visitors.
interfacebus.com [home page] -15.68% in visitors.
PCIexpress Interface -37.28% in visitors.
RS422 Interface -5.16% in visitors.
SerialATA Interface -7.19% in visitors.

Now these are not buses that are in decline, so there is no reason for these pages to see any decrease. Unlike the RS232 bus that is being left off newer computers; that page also showed a -12.25% decline, but I can deal with that.

So what is the deal with a graph that shows no real decrease, but many pages that seem to be in decline. Well when you check the blog listing new engineering pages, you'll see that over a hundred new pages have been added over the same time frame. So the new pages are leveling out this year even as a number of pages are seeing a reduced number of hits.

Hmm, I happen to be looking for the VME64x pin out today, my page came up first in a Google search but I decided to select the next guys page. Sure enough there's my text, plus a link back to me ~ stay off my site. Guess it's time to start looking at some of these other pages to see what's finding its way onto other websites.

Most of these guys running hardware or engineering sites aren't engineers at all, you can kind of tell by what they copy.

The point here is always spend some time searching the Internet looking for your work. I found a guy a few years back who had copied my PCIe page [and a dozen others], but I had just written the page. It had taken several hours to generate the different pin out tables in HTML, and he grabbed them in a matter of minutes.... Better stop now or I'll start naming names. Oh had two typo's I found weeks later, to bad he missed garbing the up-dates.

Why do Web Stat Counters Differ

So I was looking at the Manufacturers listing section again, just to see how that section was doing. Looks like there are 95 individual pages that make up the manufacturers section. There are 10 pages that make up the listing of companies that start with the letter 'A', as an example.

The graphic shows two examples of counters that might track incoming visitors, both counters represent page views.

The top counter is from Google Adsense and the bottom counts represents page views from Google Analytics. Both counters require JAVA Script to be enabled in the browser.

But why are the counters so different, there both from Google? The Adsense counter seems to be counting at a higher rate than the Analytics counter. Note the scales differ between each counter.
Click the graphic for a larger image.

Pages with the least amount of visitors;
Component Vendors, 'Us'.
IC Vendors, 'Ap'.
Equipment Manufacturers, 'Go'.
IC Manufacturers, 'Gen'.
These are the bottom 4, but it could be these page were only generated last year and may not be as old as some of the other pages.....

Alexa Demographics

Alexa is one of many web sites that collects data on web site visits, but with Alexa you have to run their tool bar. With out the tool bar they don't collect any data.

So they only gather data from a certain group of people, and not from every one who might visits a site. Because of this I've never worried about the over all stats they display because I know it's only from a small segment of the population. However at the same time, many people also think the opposite.

Any way here is a partial data chart from Alexa for interfacebus.com.

The green represent above average visits from a group, while red represents below average visits. So the web site brings in older male visitors that are highly educated while they are at work.

Click on the image for a larger view of the pic.

I Need More Visitors

Visitors to interfacebus.com have been flat for years. I add new pages all the time, and of course I up-date and enhance existing pages at an even faster rate [Engineering Page Additions]. What ever I do to the web site, it seems to make no difference.

Of course some of the issues that effect the website are just out of my control. A third of the website is geared toward Computer Buses. Well as time moves on more and more of those PC Buses go obsolete. For example the PCI Express bus replaced both the PCI Bus and the AGP Bus, resulting in 5 pages replacing 8 pages [when pinout tables are counted].

Many new pages added are small in size and it takes time to work them while also doing routine maintenance to the website.

----- RoHS Definition [Added in 2006, no Page Rank, and only viewed 347 times]

Sometimes a page might seem like a good idea, but I never get around to working that particular issue.

----- PON Acronyms [Added in 2007, Page Rank 2, only viewed 263 times]

Maybe I start a page because I generate a new graphic and think it will support another page.

----- Compliant Pin Description [Added in 2008, Page Rank 2, only viewed 230 times]

Adding new pages does not guarantee visitors, as they still have to compete with preexisting pages on the web.

----- Manufacturers of Waveguide Gaskets [Added in 2009, no Page Rank, only viewed 71 times]

New pages also need three months before they receive a page rank from Google, if they ever get one at all. A Page Rank helps the page show up on page one of the search results.

Of course I have no control how my pages show up in the search results. The rise and fall based on an algorithm running on some computer. This page lost 50% of its visits in mid 2008 and stayed there.

----- Wire Insulation Color Codes [Added pre 2006, Page Rank 2]

FYI

The attached graphic is from Google Trends showing the Internet search history for the term AGP. Click the graphic for a larger image. The AGP interface [Accelerated Graphics Port] is the obsolete PC motherboard video bus which was replaced by the PCI Express [PCIe] bus.

Web Site StatCounter

Here is the latest webStats for interfacebus.com. You'll note that there has been a bit of an improvement over the last few months.

Visits are up over the last year, that includes Unique visits and Total Visits. Normally the first of the year shows an increase in site visits, or the increase could be due to work done to the web site during the end of last year.

The increase in Page Visits tells me that the site Bounce Rate is decreasing. Which means that people are checking out more pages per visits ~ which is always a good thing.

However; over the last 3 and a half years these lines are flat ~ after 3 years of work.........

Page Rank & Listing of Manufacturers

One of the sections of the site is a listing of Manufacturers in alphabetic order, instead of by topic. The alphabetic listing of Manufacturers spans 94 pages. The section received 1,255 pageviews in the last 30 days, and 8,205 page views this year. The alphabetic index page, which also happens to be letter 'A' receives most of the hits, and the rest are spread over the alphabet.

The section help people find companies when they might not be sure of the name, and it indicates who may have purchased the company if it changed hands. It's a lot of work keeping all the links active and valid, but it does seem to serve a purpose.

Any way, several months ago I broke up many of the pages to make them smaller, reducing the number of links per page. The new pages started with a zero page rank, which is normal, but they still have not recovered. Worse still, I checked one page today and the content didn't even show up during a search ~ meaning the page was not even indexed by Google.

Pages with a zero page rank:
Engineering Manufacturers starting with 'Amd'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Asq'. links checked ~ no issues..
Engineering Manufacturers starting with 'Be'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Bl'. links checked ~ no issues..
Engineering Manufacturers starting with 'Br'. links checked ~ no issues..
Engineering Manufacturers starting with 'Cm'. links checked ~ no issues..
Engineering Manufacturers starting with 'Con'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Cp'. links checked ~ no issues..
Engineering Manufacturers starting with 'Dem'. links checked ~ no issues..
Engineering Manufacturers starting with 'Ec'. html code updated,links checked ~ no issues..
Engineering Manufacturers starting with 'El'. links checked ~ no issues..
Engineering Manufacturers starting with 'Em'. links checked ~ no issues..
Engineering Manufacturers starting with 'Fm'. links checked ~ no issues..
Engineering Manufacturers starting with 'G'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Gen'. links checked ~ no issues..
Engineering Manufacturers starting with 'Gl'. links checked ~ no issues..
Engineering Manufacturers starting with 'Go'.
Engineering Manufacturers starting with 'Hb'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Hj'.
Engineering Manufacturers starting with 'Il'.
Engineering Manufacturers starting with 'In'.
Engineering Manufacturers starting with 'Int'.
Engineering Manufacturers starting with 'Intel'.
Engineering Manufacturers starting with 'Jd'.
Engineering Manufacturers starting with 'Ke'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Ki'.
Engineering Manufacturers starting with 'L'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Mo'.
Engineering Manufacturers starting with 'Nik'.
Engineering Manufacturers starting with 'Oo'.
Engineering Manufacturers starting with 'P'. html code updated, links checked ~ One bad link..
Engineering Manufacturers starting with 'Po'.
Engineering Manufacturers starting with 'Ram'.
Engineering Manufacturers starting with 'Ren'.
Engineering Manufacturers starting with 'Sen'.
Engineering Manufacturers starting with 'Sig'.
Engineering Manufacturers starting with 'Sk'. html code updated, links checked ~ One bad link..
Engineering Manufacturers starting with 'Sp'.
Engineering Manufacturers starting with 'Tb'. html code updated, links checked ~ One bad link..
Engineering Manufacturers starting with 'Th'.
Engineering Manufacturers starting with 'Us'.
Engineering Manufacturers starting with 'V'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Ve'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Vi'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Vm'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Wi'. html code updated, links checked ~ no issues..
Engineering Manufacturers starting with 'Z'. html code updated, links checked ~ no issues..

Wow, 47 pages with a Google page Rank issue ~ that's not good. But what is the problem with the pages?

I did notice that the entire section declares xhtml 'strict' coding, so I updated a few to 'transitional' coding today. The pages will validate declaring 'transitional' coding, but that can't be the issue or they would all have a zero page rank.

Any non '.com' web site link has a "rel = nofollow" in the link so even if the link were bad Goggle would ignore it. The 'rel=no follow' must be on 20% of the links, making them save regardless.

I'm running Xenu tonight to check the site for bad links, but that program will not indicate if some other site has taken over an address ~ so I'll have to hand check each link to insure the page opens to the site listed. Xenu reports 8 bad links (0.13%), with 5937 good url's.

I hand checked a few pages, see the text above, but I don't see any issues. I don't mind so much about losing the page rank, but I would like to have the data searchable. I could be missing a number of visitors because the page are in Google's supplemental listing.

Engineering Dictionary

I took a look at how a section of the web site was doing, in this case the Electronic Dictionary section.

Here are the page views by month:




Jan: 8,254
Feb: 8,046
Mar: 8,814
Apr: 8,584
May: 7,974
Jun: 7,183
Jul: 6,707
Aug: 7,427
Sep: 9,270
Oct: 11,269

I can't tell yet if the increase in page impressions is due to adding additional pages to that section, or what. This month has 5,914 page views so far from about 279 different pages. I didn't take the time to look up how many pages made of the Glossaries section in previous months.

Update [9/23/09]; I'm not really sure what I'm counting here? There is a very small section that uses the term 'dictionary' in the address but that only amounts to a few web pages. A much larger section using the term 'glossary' in the web address is out there, but those pages do not add up to the numbers listed above either [as in Engineering Glossary].

So I no longer know what these numbers represent, maybe the addition of all the different dictionary sections? I really don't know! lets disregard this posting from now on........ Editor

Oh, the graphic is copied from the Dictionary of Capacitor Terms..