For even those companies who are most advanced in online testing & optimization, one large part of their sites usually remains relatively untouched: customer support.  I’ve been trying to figure out why the response is usually lackluster at best when I suggest running tests on customer care and knowledgebase pages.  One reason seems to be the lack of ownership within marketing.  Ever heard of a customer support marketer?

Another reason may be because it’s harder to quantify the ROI.  There’s no straightforward way to measure how effective a test is.  We could look at how many people don’t go to the phone numbers page, from which we would extrapolate that they found the answer online.  In conjunction with that, we could look at the number of page views to understand how deep within the bowels of FAQs and articles the customer had to navigate.  Does that really tell us if what we’re changing on the site is working though?

Let’s say we then track them into the front-line of phone support: IVR (Interactive Voice Response), the automated phone tree that we are all too well aware of.   That could be our negative conversion event.  But the big dollars in customer support reside in the call center.  Most companies would be perfectly happy if customers could service themselves through the website and IVR.  So perhaps IVR is the first stage in the funnel, and live customer support is the second.  As someone who lived and breathed call center metrics and performance in a previous life, I am still blown away by just how much every interaction with a call center agent costs, and how much can be saved by reducing AHT (average handle time) by even one second.  We’re talking a savings of millions of dollars for any enterprise company.

If those figures aren’t enough to get people to start testing, what about the straightforward goal of retention?  Retention is not just about sending out emails to existing customer lists in order to upsell.  As somebody who sells software-as-a-service on a subscription basis, I know that a large part of my job resides in keeping my clients happy.  Actually, it’s about more than just being happy.  I want them to be continually delighted by how much testing & optimization improves their work, whether it’s in the implementation, the test design, the strategy and execution, or the stellar results they get to evangelize.   Traditional customer support can be that as well.  It’s not most of the time, but hey, when it’s good it’s great.

The customer support section is a perfect place to test because your customers are most likely trying to shout the goal of their visits at the top of their lungs.  Look at what brought them to your site in the first place.  Did they do a search for a specific product in conjunction with “help”?  Probably safe to assume they own it.  Are you showing the most read articles for that product on the page then?  Or better yet, are you delivering the answer to their question upon landing?   If you can provide the answer, then you have a huge opportunity to cross-sell and upsell on top of it.  For example, maybe I’m looking for tips on how to extend the battery life on my laptop.  If you can take me to a page with suggestions, I might also be enticed by the extended-life battery you’re recommending for me.  Additionally, you could also profile me as this specific product owner so that you can target and sell me more effectively once I go back to the shopping side of the site.

My guess is that most people are thinking that these ideas would be nice if marketing and operations talked to each other, but right now they’re on different sides of the house.  I’m thinking that it might be time for operations to get their own set of web marketers…