Five Principles of Web Analytics: Know & Sow For Profits

How to get the most out of the top 20% of your market.

June 23, 2007
Found in: Articles : Editorial : Web Site Marketing & Strategy :

You’ve probably heard about the Pareto Principle, more commonly called the "80-20 rule". This principle states that, "for many events, 80% of the effects comes from 20% of the causes." For example, 80% percent of your productivity comes from 20% of your effort and 80% of your profit comes from 20% of your clients and 80% of your sales come from 20% of your market share.

Obviously, to get the most out of any web site, you need to know as much as you can about the lucrative 20%, and make the best use out of that data.

Here are five principles of web site analytics and measurement that will help you make the most of your best 20%.

1. Hits vs. Page Views: Tracking The Wrong Data Can Be Misleading

   In the early days of the commercial Internet, circa 1996, the technical folks were in charge and created a way to get qualitative data on web site usage via web server logs or the record of requests for and delivery of files. The focus was on technical data such as:
   

   • Hits - Requests for a file on the server.
     
   • Kbytes - Number of 1024 byte, or kilobyte, chunks delivered to users.
     
   • Files - Distinct files delivered to users.
     
   • Pages - Distinct files, specifically HTML, delivered to users.
   • Sites - Distinct other web sites sending traffic to yours. Also known as a "referrer" or more commonly known as the mispelled "referer."
     
   • Visits - Number of user "sessions" recorded.
     

   Here are some examples of what this data looked like using a log analyzer called Webalizer. This tool is still in common use today and is a great first step in getting an understanding on whether your web site is being utilized.
   

   Overview: Click to Zoom

   Detail: Click to Zoom

   One significant drawback to this method is the vague nature of the terms and what they mean. How is a 'hit' valuable to your business? Should I be excited about a high number of 'visits?' Could there be misguided excitement and therefore mislead decisions?
   

   This gets to our header for this section, "hits vs. page views." The tendency is to look to the high numbers and count them as successes. In a race, the goal is not to count the number of strides but to gain a favorable position at the finish line and ultimately to win. Since a 'hit' is a request for a single file and since a given 'page view' or user viewing a singe web page, could generate hundreds of hits, the number becomes meaningless. In this scenario thousands of hits could be created from just a few page views which could have been from a single user.

   Is this worthy to be branded success?

   No, not yet anyway.
   

   How This Relates To Real, Bottom Line Value.

   Understanding whether your web site is being used is a great first step. However, due to increased costs related to the maturity and growing complexity of the Internet, a web site's central role in sales and marketing, and the rise of the search engine as the gateway to future prospects, more detailed questions require answers to justify the resources given to modern web solutions.

   Have you ever had any of these questions?
   

   • How many first time visitors did I have this month?
   • How long did users stay on my web site?
   • Once they got to my site, where did they spend most of their time?
   • If I received any users via a search engine, what words did they use to find me?

   Or even better:
   

   • How many users submitted my contact form today?
   • Of the users who came via google.com this month, how many of them filled out my offer form?
   • How much money did the web site make me this month?

   These questions require focusing attention on the right data for your situation. Today there are hundreds of markers or key performance indicators (KPI) measured but here are a few commonly used ones:
   

   • Page Views - The number of actual web site pages that have been viewed.
     
   • Repeat Visitors - The number of visitors who have visited two or more times.
     
   • Referring Sites - The other web sites that have sent users your way.
     
   • Entry/Exit Pages - The pages that users start and end their time on your site.
     
   • Average Duration Spent On Site - The average amount of time users spend in total on your site.
     
   Notice 'hits' is not on the list as it is not useful in measuring success yet I hear many people tout this high number as though it is important. Additionally, this information is now available in many different time periods such as daily, weekly, monthly, and quarterly besides just monthly and sometimes only lifetime, or from the time tracking was started or the time period of the analyzed log file which makes gathering intelligence challenging.
   
   

   I also wanted to highlight some very powerful markers that can add significant intelligence to a web marketing campaign, especially when it comes to the vital use of search engine marketing:
   

   • Search Terms - The terms or keywords users are finding your site with.
     
   • Pages By Search Engine - The specific pages that are being found by specific search engines.
     
   • Pages By Search Term - The specific pages that users are finding when using specific terms or keywords.
     

   Here are some examples of what this data looks like in the analytics service we offer our web solution clients. We believe it is vital to have better data if we are going to answer these important questions. We go into a little more detail on some of these aspects below and will certainly follow up with more here in the future.

   Web Analytics Examples

   Traffic History. Click to Zoom

   Page Views. Click to Zoom

   Search Engines. Click to Zoom

   Latest Visitors. Click to Zoom

   
   

   
   

2. Three Types Of Visitors And What They Tell You About Your Web Traffic

   In order to really grasp the effectiveness of your web and email marketing campaigns, you can't simply measure the traffic to your web site. You need to understand the kinds of visitors you're attracting, and why they were brought to your page.
   

   As a baseline, you need to be able to separate out your unique visitors and repeat visitors. Here's an example.
   
   Suppose you look at your analytics and you see that your web page got 1,000 views this week. The amount of unique and repeat views in this number will lead to different interpretations.
   
   Let's say they were all unique views. This means that each view represented a single individual who came to your web site once. You attracted 1,000 new people to that web page, and if this is a high number you know that your advertising, Search Engine Marketing/Optimization (SEM/SEO) and other efforts to build traffic are paying off.
   
   But this isn't always good news. If you're trying to create a social network where people bookmark your site and return to it often, a large proportion of unique views is bad news. These visitors only came once, and didn't feel the desire to return during the time period selected.
   
   Now let's look at the opposite scenario. Suppose that out of 1,000 views, 800 were repeat visitors. This suggests you only attracted a few hundred visitors at best, a much smaller number. It suggests you need to find ways to bring in more traffic.
   
   On the other hand, a large number of repeat views shows your web site has something to offer. People are coming back.
   
   This can provide a lot of important clues. Repeat views of a product description might indicate people are thinking about buying, but are indecisive. You could respond by posting additional information, or emailing a special promotion to convince these red hot prospects. If the visitors are viewing an article or other source of information, you now know what's popular and you can give them more of the same.
   
   A third variation on web traffic is campaign visitors. These are visitors who click through to your site in response to a specific campaign. Somewhere, somehow, in a book or a seminar or an email or an ad campaign you asked them to go to a campaign related web address, and they did.
   Obviously, this is an important measure of the success of your campaign and a way to differentiate traffic. But you can learn more even beyond this.

   If campaign visitors tend to convert better than visitors from other sources, this could be a sign that your web pages alone don't convert, and you may want to tweak your copy. If the campaign conversion is lower, it suggests your campaign isn’t bringing in the best prospects, and perhaps you should target a narrower niche.
   
   

3. Duration And Navigation Path

   Before you’ve really made the most out of your web analytics, you need to interpret the length of time spent at your site, a particular page, and the navigation path the user followed.
   
   User visits with a long duration show that your prospects are giving a lot of thought and attention to the content on your page. Sometimes this might be a fluke (they took a coffee break while your page was on their browser) and other times it might happen because you have a lot of content for them to get through.
   
   But consistently high durations show sincere interest and thoughtfulness. About the only time this could be a bad sign is on a conversion page such as a contact form, where it could suggest indecisiveness or even trouble filling out the form!
   
   Along with measuring how long a visitor stays on a page, you need to know where they clicked from, and where they go after they arrive at your site. This navigational path gives you a lot of clues on how to improve your site: If visitors don't click through the link that you want them to follow, you may need to revise your copy. If they hit the back button a lot, there might be some important information on another page that isn’t clearly spelled out.
   
   Knowing where visitors were before clicking though to a page also gives you valuable insight into ways you can drive traffic to that page.
   
   

4. Entry And Exit Pages
   
   Your visitors' navigation paths are only half of the equation. To maximize your web site marketing, you should understand where and how they enter and exit your site.
   
   The default entry page is usually your home page but can also be a specifically tailored landing page. This page welcomes the visitor, delivers on something promised in a web search or campaign, and directs your visitor to click somewhere else next. But not everybody enters your site at the appropriate landing page.

   Some visitors may go directly to a product page, or a popular article or blog entry. Maybe a random, seemingly insignificant page gets a lot of visits because it happens to be optimized well for certain keywords.
   
   All this entry page data has important implications for the way you run your search engine marketing. It can also give you clues about new ways to drive more traffic to your site.
   
   If entry pages help you uncover hidden opportunities, exit pages are a good way to search for problems. An exit page is the last page a visitor views before leaving your site. If it happens to be a "Thank-you for your order" page, then you might be satisfied with your navigation. Or you could add something to the page that will encourage future visits and sales.
   
   If people are leaving your site from other pages, you need to examine them carefully for changes. Is your content too long, irrelevant, offensive, or boring? Do you need a more aggressive prompt or offer to have viewers click through to another page? Can you make the navigation bar more prominent or usable?
   
   

5. Action Points And Transactions
   
   Finally, you should understand how to track your visitors' activity at key action points. Whenever they have a chance to do something -- watch a video, download a free report, sign up for a newsletter, or make an appointment -- your visitor will either do it or not. And then they'll do something else after they've made their choice.
   
   You need to monitor these choices. If too few visitors are taking the desired action, you have to reevaluate your copy, your content, and whether or not it's realistic to expect them to take action.
   
   Transactions offer a special case. If a lot of shopping carts are abandoned, is this a usability issue or a marketing problem? Are clients more likely to make a transaction after viewing certain web pages?
   

Putting Together The Five Principles Of web site Analytics

As you can see, maximizing the most productive 20% of anything is both an art and a science. The science is collecting the data and measuring the numbers. The art is interpreting the data and making changes based on what you learn.

If you want to get the most out of your web site marketing, you'll need to do both. But by mastering the five principles we've described here, you'll have a framework that will make the job a lot easier.

Make Web Analytics Work For You

There are two factors that can make or break your web site investment, getting the right data and interpreting it correctly. Enthusiast provides comprehensive web solutions focused on results. Start a conversation today and receive our FREE report titled "5 Ways To Maximize Free Web Traffic Today" with no obligation."

* This article is part of the "Measurement & Analytics" topic from the Enthusiast Editorial Calendar. You can get more of what you want and help us by suggesting a topic or asking a question.