Regular readers of The SSL Blog will be familiar with the wealth of research indicating that the presence of a green address bar on a Web site causes an increase in transactions among visitors who see them. With over 60% of site visitors on EV-aware browsers today it's straightforward to calculate the expected impact of those green address bars on your KPIs, assuming that average transaction value (ATV) remains the same.
Assuming that ATV remains the same.
But can we make that assumption? There is research indicating that ATV goes up among those who see green address bars. I refer to two particular studies. The first is research conducted on its own customer base by CanadaDrugs.com. As the name implies, CanadaDrugs.com is a pure-play online drug store servicing the Canadian market. CanadaDrugs.com compared abandonment between visitors with EV-aware and older browsers and saw that the number of tickets sold increased by 27% when customers saw green address bars over when they did not. Okay, that jives with what many other businesses have seen. CanadaDrugs.com looked into another interesting question as well, however. The company compared the average ticket size between the two groups and found that the EV purchasers had tickets that were an astounding 30% higher on the average. In addition to the 27% increase in transactions completed.
Layer the two on top of each other, and you get an actual contribution to the business of 65% (1.27 x 1.3). A visitor who comes in with an EV-compatible browser (an expected three out of five visitors) is worth on the average 65% more in sales than that same visitor would be if the EV SSL Certificate were not in place.
The second piece of research investigates this behavior more closely. In June, 2007 Carnegie Mellon researchers Janice Tsai, Serge Egelman, et al published the results of their research on privacy and price sensitivity. Test subjects were given a budget to make online purchases using their own personal information (e.g. credit card, shipping address), and whatever money out of their budgets they did not spend they got to keep. Subjects were asked to search for products online using a tool that rated the privacy protection of each site the subjects might visit. In other words, the test subjects were making real decisions about their own private information and using their own money.
Test subjects did not simply go for the lowest price product. They were willing to spend a higher amount on the sites with good privacy protection ratings than they were on the sites with low ratings. That makes sense. The test subjects essentially were paying a relatively small amount of money as insurance against something bad happening with their private information.
It's tempting to attribute this same explanation to the CanadaDrugs.com finding, and I do believe it's relevant, but we need to modify our thinking a little. The Carnegie Mellon research saw an increase in average selling price (ASP) when visible privacy assurance was in place. Subjects were willing to pay more for the same item because they felt safe. The CanadaDrug.com research saw an increase in average ticket value (ATV) when visible privacy assurance was in place. Subjects still paid the same amount for their products - ASP was the same - but in this case they bought more or higher priced items.
So what else is going on here? I hypothesize that we're seeing the effect of the fact that not all online sales tickets are the same. Not all shopping carts are equally full. Some people go to CanadaDrugs.com to purchase only one relatively inexpensive item. Others go there to buy lots of items or items that cost a lot more money. I can't put my finger on it at the moment (I'll look around and try to find it), but I have seen research in the past indicating that shoppers perceive their personal risk to be higher with online credit card purchases when the price is higher. Your average online shopper believes she's at more risk when making a $1000 credit card purchase than a $10 credit card purchase. This perception in many ways is invalid in that both the risk and consequence of credit card theft are identical for purchases of all sizes. Nonetheless, people are more skittish when they have a large ticket.
I suspect that's what's going on with CanadaDrugs.com. The average propensity to abandon a transaction increases with ticket size. So people with fuller baskets in the aggregate are abandoning more carts than those with emptier baskets. Now the site does something to display its premium security and increase customer confidence, and as a result users abandon at lower rates. Since abandonment is higher among the high ticket carts, more of those customers finish their transactions, and ATV goes up.