Learning how our web site works is an important task. We’ve seen some simple examples in previous steps (visits, page views, and unique visitors) as well as determining how long someone visited your site, and which was the last page they viewed. We’ve shown how some pages are viewed more than others, and where people exit your web site. We’ve even shown what pages people come to your web site.
What we will look at now is how did they get to your web site. Before they were on your web site, odds are they were somewhere else. Knowing this tidbit of information changes everything, and starts to allow us bring things together.
In the past articles I’ve had a decidedly split view, why this matters, why this doesn’t matter. This is going to be one sided - this matters. And the follow up where your (quality) visitors come from, will matter even more as we show you how the metrics from the previous articles start to fold in with where your visitors come from.
You will get visitors from all over the Internet. Web sites you don’t know about link to you, and such. However, all of these methods/places can be grouped into the following categories.
- Search Engine (organic) - this is where someone types in a question, and you site was on the list of answers. Depending upon the phrase entered, you may have little (rarely) or lots of competition.
- Search Engine (paid inclusion/pay-per-click) - You have purchased key words/phrases, and when someone enters a phrase, your ad comes up. That person clicks on the ad, instead of an organic result, you get charged a fee, and they go to your site.
- Online Ad Network - other networks beside those run from search engines exist, and you have purchased ads to be placed on websites based upon some criteria, and users came to your site from these ads.
- Another Website - another website, be it a standard site, a blog, a forum, etc. has your.
- Direct Access/Bookmark - someone typed in your web address or had previously bookmarked your site. This number is often a higher than it should be as sometimes the information used to pass where the visitor came from, isn’t passed. In that case, the Direct Access gets artificially inflated.
The question of course is, why is it this that matters? Here are how we look using the previous metrics to start to determine relevancy, and in how this metric matters.
Lets say you have a small business. And you have a few friends who like to refer their friends to your business. We all know that all referrals are not equal, and someone is going to send you better referrals than others. But how do you know which is which.
Obviously, the one that leads to the most sales is sending you the best referrals. But can you find out why that friend gives you the best referrals? Or better yet why?
You see, search engines, other sites that link to you, and your banner ads are your “friends”. They all want to send you visitors, just some are better at it than others. And just because someone refers a lot of people, doesn’t mean they are sending you the right visitors.
Lets say we look at some on-line ads you are using. You know that you are getting leads from them, sometimes 1000 a day. However, the users leave quickly after viewing the page. If you can segment your viewers and determine that 80% of the visitors that come to your site from this source “bounce” then you know you have a problem you have to address. You don’t know exactly why, maybe the leads/visitors aren’t qualified, or maybe the landing page (the page the visitor sees when clicking on the ad) is ineffective, but now you have something to start looking at to determine what to fix.
On the other hand, you have 1000 people coming from another web site, maybe Google Ad-Words and Yahoo’s pay-per-click. You can see from there that you have a bounce rate of 40%, and they tend to stay for 3 or 4 pages, by then 60-80% of people have left your site. Now we can look at a different set of problems. Could they not find how to buy? Did they get “fatigue” where they got tried of going through the pages? Now we can look at the exit pages, do they move from page to page as expected, or do they bounce around unexpectedly?
As you step through these numbers, I’m sure you will see that not all sources are equal. Some might result in lots of page views, but few conversions (sales, or requests for more information). This is why visitors, page views, and time on site are not necessarily good metrics. You have to learn what type of lead from a source is your best type.
Additionally you can start to segment your search engine data, be it from normal, organic, searches, or from paid search results. Now you can start to see what words and phrases people are using to find you, and how effective you are at converting those people into buyers. This is where your metric data starts to become real important.
Knowing what keywords and key phrases people are using to find your site when dealing with search engines. You might find that the phrases you would use, and you have optimized your site for, are not being used. Is this because there is too much competition, or because consumers use different words than expert/owners. It can also reveal when a phrase is ineffective, because it yields too low a conversion rate.
Knowing how many people repeat visit your site, and/or bookmark you site and directly access it indicates how valuable other’s feel your web site is. Likewise, if you find that most people visit from a given source once, but never return, you can start to see that they don’t place a high value on your site.
You may not be able to determine why people are leaving your site, but you can start to develop some good theories. Just start looking at it from a users point of view, and see what if anything you have to do to enhance your site to increase those conversions.
Photo by: 
