Friday, 4 February 2011

Lies, Damned Lies, And ...

... Statistics !!

Having trouble sleeping ?

Reading this will probably get you into sleep mode, real fast. It's just some thoughts, and observations, about blogging and traffic statistics. However, if you are a 'blogger', and this sort of stuff interests you, read on.

If you are having trouble sleeping, read on also. It will probably help.

Before going further, let's examine what a blog is. The original term was 'web log' which quickly became shortened, and popularised, as blog. Blogs are typically a venue for commentary, information, or some are simply more in the form of a personal diary.

If you look at some of the web ranking sites, there are many blogs which attract a huge following. Some for good reason, some not.

When I started this blog, just over four months back, I was mostly doing so for a place to put images, with a little text, which I could then link back to from some of the photography forums I participate in.

One of the problems with some of the larger photography forums is that information can get lost in the volume, and I'm not just talking about post quantity.

I also wanted a place where I could talk a little about cameras I use, and also about some post processing things I have learned over the years, and which I am still learning about.

Just my thoughts, my experiences and part of my daily activity.

Well it didn't take long before it started to attract a very diverse audience and have many links from other photography sites. I still puzzle over why a site about dieting links in though.

The blog also very quickly started to get a lot of 'hits' each day. How many, is not so important. There is more to it than just 'hits'.

To give you a rough idea though, I posted a post processing tip today and in just the first hour alone it had over a hundred visits. As yet, I know nothing about whether anyone found the information useful.

Now blogs are a little different to other web sites. If the blog is not intended to be a source of revenue, then getting too concerned about the numbers is pretty pointless. After all, it's just a blog.

However, a blog can generate revenue by providing advertising links. It's a different story then.

About a month after I started my blog, and it was clear that the traffic was climbing very quickly, a friend asked why I didn't have advertisements. At the time, I stated that if my traffic rate doubled, from it's then position, I would look at it.

Well, it's now many, many times higher than it was then, and I may introduce some advertising later, but also may not. Never say never but, for now, it's just a blog. Simply my space to share things with like minded folk.

Anyway, to get back to the story - Statistics are, in a simple sense, where we collect data and then interpret what that data means.

Data, in a numeric sense, is simple - I have a quantity of ten apples. Eight are red, two are green. Of the ten, four are ripe.

So quite quickly, and simply, I can 'interpret that data' as 80% of my apples are red and that 40% are ready to eat.

It tells me nothing about other factors though. Size, taste, quality, etc. For that, I need more information. Information about quality, not just quantity.

If I give one of my ripe apples to a friend, what I would like to know is, at the very least, did they enjoy it ?

The more data we have, and the more that is fed back in the collection process, the more we are able to interpret that data. It's the same with blog traffic. I can see there are many visitors each day, but are they enjoying the experience ?

For our purposes here, the interpretation of that data is to examine the behaviour patterns of blog visitors, and how that relates to the content of the blog. What we are most interested in is the 'performance metrics' which can be used to assess what visitors do, and how they react.

Performance metrics are simply tools which allow us to analyse what visitors do. It can become clear what they enjoy, and what they don't.

They enable us to work out whether the blog content is worthwhile, or not.

So, what has all this to do with lies and statistics ?  Let's leave the 'lies' out of it, but talk instead about statistics and, particularly, behavioural statistics.

I am often amused by statistics published about blogs. Many people simply publish traffic numbers. That is, how many visits, page views, etc. Boasting about having thousands of visitors, without telling the whole story, is really quite meaningless.

Among other things, what is meaningful as quality measurements are - How many visitors enter and stay, how long they stay, how far they travel through the site, how many come back, and how many new visitors come to the site.

So, for the purposes here, let's ignore the actual quantity measurements, and take a little more in-depth look at some of our 'quality' measurements.

This is by no means an exhaustive list, but some of these things are measured as :
  • Bounce Rate
  • Time On Site
  • Average Page Views
  • Returning Visitors
  • New Visitors

Bounce Rate

First off is 'Bounce Rate'. What this measures is when a visitor enters the site, looks at a page and then exits before a preset time, or before going further into the site. Usually this is referred to as an instant exit. It is not 'exit rate' though. At some point, all visitors will exit the site.

This metric can be applied to the main page, or for any single page. In fact, most web site analytical tools break down this measurement for all pages on the site.

Note : In the data I list below, I am simply using the measurements provided by Google Analytics.

In effect, what this means is that if a site had one hundred visitors each month, and all only viewed the page they entered at and said "This is not where I want to be" and left, then effectively the sites traffic is zero. That's right zero.

Imagine if you ran a restaurant. Each day one hundred people stop at the door but all just stick their head inside for a few seconds, don't see anything that takes their fancy, and leave. Then your 'bounce rate' is effectively 100%. In reality, you'd be out of business very quickly.

Similarly, a site with one hundred visitors per month and a bounce rate of 70%, really only has an effective visitor rate of thirty. A site with a visitor rate of fifty visitors per month, and a bounce rate of 20%, effectively has forty visitors per month.

So, examining bounce rate and keeping it low is important.

As of the time of writing - The benchmark bounce rate for sites of a similar size to this blog is 48%. My current bounce rate is 53%. Not too bad but clearly there is room for improvement.

Time On Site

So the site has a visitor, but how long do they actually stay on the site ? It is an important measure as this also gives a clue as to how much people enjoy their visit.

For example, it took me three minutes to read down to this point, on this page. I am a speed reader but reading carefully, and ensuring I understood what I just wrote, took me three minutes.

So, if someone visited this site and I could see that for this particular page that the 'time on site' was, say, one minute, then I would easily understand that they thought it was just boring 'waffle'. And, it probably is. It is still a measurement though and conclusions can be drawn from understanding it.

As of the time of writing - The benchmark time on site rate for sites of a similar size to this blog is two minutes and twenty three seconds. My current time on site rate is three minutes and forty two seconds.

Average Page Views

Each site has a number of pages. As at the time of writing, I have over a hundred pages on this blog. When someone visits, they may do so by coming into the main blog itself, or by entering from an internet link somewhere, to a particular page. This is recorded not only as a visit, but also as a 'page view'.

If the visitor visits several pages on the blog, then that 'movement' is also recorded. If two visitors enter the blog and one views six pages, and the other four, then the total page views will be ten, however the 'average page views' will be five. Simple math. 6+4 = 10. 10/2 = 5.

As of the time of writing - The benchmark average page view for sites of a similar ranking is 3.3 pages which is the same as my current average page view rate. With over one hundred pages on the blog, it would be good to try and improve on that.

Returning Visitors

Back to our restaurant scenario. We've got the business sorted out a little better. Now, half those who come to the door are actually entering and sitting down for a meal.

The problem is that the town we have the restaurant in, has a total adult population of one hundred. Therefore, we know that our potential ongoing customer base is fifty people, as we are currently seeing half of all visitors come in to dine. The problem now is, how many will come back ? If 100% never come back, then we'll be shutting down real soon.

Again an important measurement to understand.

As of the time of writing - The benchmark new visits rate for sites of a similar size to this blog is 63%. My current new visits rate is 59%. The returning visitors rate for this blog is 41%. Clearly another area which would be good to improve upon.

New Visitors

Having content which visitors will enjoy first visit can help to promote any site. They will tell their friends, who will tell their friends, and so on.

As mentioned above, the new visits rate for this blog is currently 59%. Clearly there is growth and it would be good, based on current traffic numbers, if it stays at that rate, or greater. However, having content which attracts people to return is also important.

EyeMindSoul Statistics Analysis

If I take a quick look back and compare data from two months prior, with what is happening currently, I see the following :
  • Total visitor rate has risen by 82%
  • Absolutely unique visitor rate has risen by 109%
  • Total page views have risen by 149%
  • Average page views have risen by 37%
  • Time on site has dropped by 9.5%
  • Bounce rate has improved by 7.6%
  • New visits have risen by 11%

If I go back three months, I see the following :
  • Total visitor rate has risen by over 1,200%
  • Absolutely unique visitor rate has risen by 1,100%
  • Total page views have risen by 1,515%
  • Average page views have risen by 23%
  • Time on site has dropped by 18.3%
  • Bounce rate has improved by 8.6%
  • New visits have risen by 17%

So, all of the above is helpful for understanding visitor behaviour. It is by no means a definitive list of performance metrics surrounding the blog but, at a minimum, it can give me clues as to the blogs activity and what is/isn't worthwhile content.

As the blog is relatively new, it is really too early for me to draw any real conclusions. Suffice to say that it seems what I post, particularly regarding camera information or post processing, is being well received.


This brings us to feedback.

For this blog, if I write it, and I am the only reader then it is a little bit like my own 'secret diary'. If however others also enjoy the experience, or learn something, then it becomes more of a blog for sharing what I do, have done, and have learned. That was why I started it, and I am enjoying the experience.

I have many friends and contacts across the internet and I get a lot of personal emails about the blog. It's all positive comment, but then they are my friends.

Getting comments on a blogs content from 'strangers' is good. If people visit and find the content worthwhile, then it is good to know that.

There are many blogs out there, and there is some really great information. I am often perusing other photography blogs/sites and am continually learning. The ones I always get the most from are written by photographers, not just people who write about photography. I don't always leave comments but try to, particularly whenever I find something very useful.

When I started in photography, a very long time ago, most information came from magazines and books. The best information, however, always came from friends who were involved in the photographic profession. They had been there, done that.

One friend, in particular, helped me enormously. He just knew what to do, and how to do it, for any given photographic challenge. I lost touch with him many years back but I always feel indebted to him. So, Kerry, if on the off-chance you read this, thank you sincerely.

Anyway, if you're 'blogging', particularly about a subject such as photography, don't get fixated on the numbers of visitors/page views, etc. Instead, look at whether people actually find what you do is worthwhile.