Understanding Web Stats

As a web-based business owner it is important to understand how people view your site. When a website is created it is usually designed to put forth the information that represents your company in a stylistic way. As a business owner it is important for you to understand how visitors find your site, and how they interpret your site once they arrive. This is where understanding your web statistics is crucial.

As someone who has been in the internet marketing industry for years, I receive many calls regarding statistics. From these discussions, I have put together a few misconceptions and some tips to help you understand exactly how your website currently performs. For illustration, I will use Google Analytics when discussing web statistics, although almost everything I discuss can be applied to other web statistics systems. Google Analytics is a free and comprehensive web statistics system and I recommend using it to take advantage of its substantial feature set. Speak to your web developer about applying it to your website.

Top 4 pet peeves in web statistics

1. Website hits.

People often call and relay to me how many “hits” they have when compared to someone else in the same industry. “Hits” is a term that refers to the number of times a file has been requested from a web server. Since a web page is built using several files, including images, this number is never relevant for gauging human traffic. The most relevant information we can use to determine the amount of traffic a website receives is referred to as “Unique Visitors.” This represents the number of individuals who visit your site, with each computer being counted only once. Unique Visitors will be what we refer to as “traffic.”

2. Targeting pay-per-click locations

Pay-per-click advertising clients often ask me to better target people in Brisbane because they have noticed from their web statistics that most of their traffic originates in Brisbane and other capitals (as indicated by the larger dots in the example image). Unfortunately, Australian Internet Service Providers are not as regionalised as they are in other parts of the world, and a person in Cairns who connects to the internet will often show as coming from Brisbane.

3. Reading the data incorrectly

There are some very useful aspects of web statistics programs that can be challenging to turn into useful information, such as “Traffic Sources” and “Bounce Rate”. Traffic Sources tells you which medium (such as a search engine) your visitors use to find your website. Bounce Rate, also called Percent Exit, is indicative of your website’s content, and how people feel it relates to what they were searching for.

With Traffic Source, the important point to note is the overall percentage of visitors that view your site. This can be pivotal to illustrating many different things. I feel the best use is to gauge the cost-effectiveness of paid advertising. If you are spending 10% of your marketing budget on Google AdWords, and 20% to advertise on a local web portal, you would expect the local web portal to bring in twice the amount of traffic of AdWords (or twice the quality, if you can measure that.) If it is falling far short of your expectations, re-evaluate your marketing spend.

Bounce Rate is the number of people who leave the website while on a specific page. For example, if your homepage has a bounce rate of 25%, this means 25 out of every 100 visitors left the website after viewing the homepage. This tells you what your visitors think of the content on that page. If the majority of your website has a bounce rate of 35%, but your rates page has a bounce rate of 95%, it suggests your visitors are shopping around, your rates are too high, or too low.

4. Website designs that don’t reflect statistic history

If you redesign your website I recommend you keep track of your web statistics prior to and after the redesign. For a comparison to work you realistically need at least one year of data leading up to the redesign. After the design, compare the traffic to the previous few months and to the same months in the previous year (as seasonal trends are a large factor). This data can tell you what visitors think about the new design. Also, when meeting with your web developer to discuss the design, plan to meet a few months after the design to go over the web statistics. Sometimes a little tweak can produce a substantial difference.

Small changes can have a great effect

Working in the internet marketing industry one learns pretty quickly that small, targeted changes have a large effect. When working on a client’s account, I usually spend 75% of my time analysing data and a mere 15% actually making changes. (The other 10% is explaining what “hits” are.) I have found that in order to improve your site, you need to understand why people are not booking. Is it that there are not enough people coming to the site? Are they finding it difficult to use? Is it irrelevant to what they are searching for? These are more easily answered with data behind you.