Deep measurement
Wow, how fast weeks pass these days.
Hope you all had a great week full of good output and enjoyed these spring days.
Today I will discuss the traffic of your website, where do you receive it, who is your referrals and what’s their relation with bounce rate and conversion ratio.
Before I start check these two links for referrals and traffic definition and get back here to continue reading.
- HTTP Referrals
- Web Traffic
Now we know the definition for both traffic and referrals let’s dig and see how these two terms control somehow our conversions.
We agree that every website or webpage on the internet receives traffic through certain referrals (If you don’t receive traffic, then your website really in a bad shape and need to be optimized). That traffic to your page need to be segmented to determine where it comes from and how does it influences your conversions and I will show you how.
Let’s take a sample site that receives in a particular period (a month for example) a sum of 100,000 unique visits, digging inside this we will notice that 30% of this traffic you are receiving from search engines, 20% from other websites, and 50% are not referred (pure example for easy understanding).
What I did is I segmented the data I received simply by dividing the traffic into three groups:
- Search engines referrals
- Other websites referrals
- Not referred traffic
This was the first step of the exercise; now let’s take it step deeper. Your 30% search engines referrals are coming from 15% Google, 5% Yahoo, and 10% Live, and the bounce rate of these referrals are 20% from Google, 40% from Yahoo, and 60% from Live; are you following?
Now let’s check the other websites that are referring to my website or webpage. There are 20% from other websites, directories, blogs, social networks such as digg, facebook, sphinn, YouTube, Flickr…etc. You have the percent of each one and the bounce rate as well, getting these numbers depend on your web analytic tool you are using.
The third element here is the not referred links, which happens because someone book marked your website or entered your domain name directly in the browser, or got a link from mailshot or newsletter or probably received an invitation from a friend through an email. This kind of links is mostly affected by your offline marketing campaigns or the word of mouth or your popularity. You should know their bounce rate for each element and the conversion ratio; how is that?
For your online campaigns you may add a variable in your links that will explain where this referrals came from, add another variable to your mailshots and newsletters so you can track those referrals, for those who typed directly into the browsers; I don’t know if they can be measured (add in the comment if you know a way) but I’m sure there is a way through the log files.
Ok, enough of that talk and let’s do some business here, let’s show how that will improve our conversions and get us the cash. First what we did above was normal web analytic and data collection with minor segmentation.
We can improve that; get deeper and show every referrals conversion and check which is getting us the less bounce with more conversions and try to focus on it, also will give us an indication on our bad referrals that are sending us the less conversions and higher bounce rate and try to improve them.
From the above exercise you will know how is your online campaigns are working and how much loyal visitors you are receiving, you can see if you need search engine optimization, will also give you indication if you’re giving the right content to the users.
This was the baseline where you will know the indications and the trends from your flow of data you are receiving.
This post along with the previous two posts about bounce rate and conversion ratio are the baseline for your improvements, it’s not the end of the story it’s just the beginning.
Competitions online are huge, try to type a phrase or keyword your website about and check the number of results around the internet on that topic.
I hope I explained it clearly; let me hear your questions.