Back to main demo page
One of the jobs of today's on-line marketer is to turn all that wonderful visitor traffic into leads (or names if you prefer). We have seen techniques around funnel leakage, micro-conversion points, landing page optimization through nav removal, A/B and multi-variate testing but as a whole these techniques are passive in their approach.
At some point in the visitor's journey with your on-line content, there is a make or break moment where after the nth session or page view where the visitor has consumed enough of your content but will never exchange their contact information for your offers. You can determine the drop-off inflection point through your web analytics in terms of your average page visits after you remove your bounces and visitor who do not complete a goal (form conversion).
And the same goes for non-bounced returning sessions.
Once we know what levels to set the trigger points at, it is time to put the analytics to work and interrupt the visitor's flow by putting an offer or conversion option in front of them. To do this, we are going to use a series of web cookies to track their engagement.
In order to gather data, one of the most useful method is through cookies. To make sure the visitor has cookies turn on, the browser needs to be tested if it will store data.
Before tabbed browsing, sessions use to be simple, a new browser window mean a whole new session. Now with tabbed and private (Incognito, InPrivate and Private) browsing, the concept of a session has become much more complicated. Still the browser controls its various sessions internal and we can rely on session cookies to detect when a new session has started.
For session count, we can use a session cookie. By setting a unique value into a session cookie, and testing for the presence or absence of the cookie, we can determine if we are in a new or current session. Once we know this the session status, then we can derive all sorts of anonymous information about the visitor including:
|New vs Returning visitor||Visitor greeting|
|Session entry (first) page||Behavioural scoring|
|Session entry query string|
|Session referrer||Competition detection|
|Session Page count||proxy for content engagement|
|Lifetime (absolute) Page count||deeper proxy for content engagement|
Online attribution can be very useful in trying to sort out how to spend your resources in attracting your visitors to your website but making effective measurements is the other half of the story. I will not delve into web attribution here other than you refer you to Avinash Kaushik's Occam's Razor blog where he has been writing and educating about web analytics for over five years now.
Marketo by default via their munchkin code captures Original Referrer, Original Search Engine and Original Query String for first time visitors to provide you with initial lead attribution source. Any of the “Original” type fields can only be inserted into once and never updated. There is also a subtlety to munchkin code where not all of the “Original” parameters are populated at the same time. Basically, if the entry condition satisfies the parameter collection only then the field is populated.
Say the first visit happens using a paid banner ad, then the Marketo field Original Referrer will be populated with the source (referrer) site's URL but none of the other fields will change. On the second visit, the prospect remembers our ad but uses Google to find the site by typing in a query and is directed via a SERP link, now Original Search Engine and Original Query String will be populated but Original Referrer will NOT be updated.
Within Marketo, there are five (5) Smart List elements you can use for original and session attribution. For original attribution you can use the following:
Note: Salesforce's integration with AdWords (SFGA) has been depreciated as of the Spring '12 release and support for all users will end Q1 2013. Additionally, SFGA is not available to new customers as of Feb 12 2012
Attribution and counting code
Now that we have some data on how many pages and how many sessions the visitor has made to our site, you can choose when to interrupt there flow and present them with an offer that will entice them to convert by filling in a form. You could redirect the on to page containing a form but in this day and age my preference is to use a lightbox type method.
Using the web analytic data at the vary beginning of this article you have a good guess as to when you interrupt your visitor. In this specific case I would setup the lightbox to fire on either two conditions:
Below is a code sample to fire a lightbox (fancybox in this case) interrupting the visitor.
Interruption code and lightbox
If you have just arrive on this page, then you will see very little happening, but if you close your browser and return OR view 2 more pages then return to this page, I use this technique to show you how it could work. In the case of Marketo, the fancybox can be setup to use an iframe very easily so you can insert your offer.
Use the button below to force the lightbox to show
Back to main demo page