The ROI Revolution Blog

Google Analytics and the Extended Sales Cycle

June 9, 2006

Some organizations have extended sales cycles spanning multiple visits to the site, with unusually large amounts of back-and-forth between the site and the user. Typically, knowing what drove a visitor to the site immediately prior to their conversion is what’s important to a marketing team, but what if we want to keep track of what initially led the visitor to the site? In these cases, the initial referrer is more valuable than what got them to the site the second, third, and final time.
It provides essential intelligence about attrition rate, customer loyalty, and ROI.

In our experience with such clients, we have encountered some very intriguing challenges while implementing Google Analytics.
One of these clients, a leading physician in his field who performs elective operations on patients who travel from all over North America to see him, receives a large amount of paid traffic from many of the major search engines.

Sounds like every other website, doesn’t it? Well, here’s the rub. The client’s sales cycle is long enough that visitors are hitting the site multiple times before their final conversion. There are a number of steps along the way where our referral information can get overwritten. Let’s take a quick look at the entire process.

Visitor X comes to the site from PPC Ad Y and browses around. She fills out a preliminary form that gives the physician’s office contact information, including her email address. She then receives an immediate confirmation email that contains a number of links to medical articles on the site. Visitor X did not notice these articles on her first foray through the site, so she clicks on a few of the links and reads up on the procedure she’s thinking of having.

Over the next few days, while waiting for the doctor’s staff to get back to her, she heads back to the site for more research. She adds the homepage to her bookmarks, this time, so she doesn’t forget the URL.

After about a week, she receives a phone call from the staff. She’s ready to go through with the procedure, she discusses payment details, and they schedule her appointment. The only thing left is for Visitor X to fill out a form on the website, providing key medical history and giving the physician permission to access medical records. The representative on the telephone tells her to expect an email with this vital next action.

She then receives the email, which instructs her to click a link to fill out the form. She does so, fills it out, and submits. She’s ready for the procedure, she’s made her payment, and this form submission is the ultimate conversion that our client has been waiting for.

Can anyone see where our vital first contact metric gets lost? There are at least four chances where Visitor X could feasibly overwrite the initial referral: PPC Ad Y:

  1. Instant confirmation email with links to medical articles on the site
  2. Direct visit to the site for research while she waits
  3. Bookmark to the site
  4. Email from staff directing her to physician permission form

By default, Google Analytics does not overwrite a visit from a UTM-tagged link with a direct visit to the site (i.e., address typed into a browser, or bookmark accessed). It will overwrite with any tagged link, any referral from another website, or any search query. So, in our example, if both PPC Ad Y and the instant confirmation email were tagged, PPC Ad Y gets overwritten, and the eventual conversion will be attributed to the instant confirmation email.

So, we don’t tag the emails. Easy enough, right? The only problem here lies with the growing popularity of web-based email. It turns out that Visitor X uses Yahoo! Mail. So if we remove the UTM tags from the instant confirmation email, everyone using a client-side email client is fine, but Visitor X‘s initial referral, from the PPC ad, gets overwritten, and her conversion is now attributed to something like "us.f326.mail.yahoo.com".

There’s a way around all of this, though. We tag the email links, and any other referral links, with the utm_nooverride parameter. Simply tack this onto the end of your links like so:

<a href="http://www.roirevolution.com/blog?
utm_nooverride=1">Click here!</a>

The 1 tells Google Analytics to turn on nooverride, which ensures that this referral does not overwrite any existing referral information. You can use this on email links, referrals from external sites (just ask politely if the site owners don’t mind adding the parameter), and tagged PPC destination URLs. Any links you have control over, basically.

For the majority of sites, it’s not absolutely necessary to attribute only the very first referral to the sale. Some prefer to see the latest action that a visitor made to reach their site before a conversion. With Google Analytics and the utm_nooverride function, however, you can customize this metric to your own specifications.

Take the next step. Ecommerce retailers spending at least $5,000/month in AdWords qualify for a free 20-minute AdWords Diagnostic Checkup.

© 2002-2014 ROI Revolution, Inc. All rights reserved.