Evan Schuman's StorefrontBacktalk

The very last thing that an E-Commerce customer sees?and it’s often what will influence whether the customer clicks on that final purchase button?is the checkout process.

Fitting an E-Commerce of such importance, it’s also the segment that is the most challenging for IT. Beyond the listing of the intended purchased items, it needs to navigate discounts, state tax, shipping charges, shipping arrangements and payment processing, along with inventory access (both to verify that the product is available to ship and to make sure that the inventory is reduced once the order is finalized). Remembering that each of those requires interaction with a different database and sometimes external companies, such as a retailer that needs to schedule shipments through FedEx and payments through various processors depending on the selected payment method.

It’s no wonder that so many purchases are lost during those final seconds, as some glitch kicks in from any of those places.

But given how important numbers are to this process, it’s sad that numerical comparisons on various final-stage services often fall so short.

For example, I noticed this interesting Bloomberg story about how well EBay’s PayPal is faring against Google Checkout and *sighed* at how weak a foundation it was built on.

First, the story seems to be based on the statement that “In recent surveys, the world’s largest auctioneer found that less than one out of five users of Google’s Checkout online payment service was happy with it.” But it doesn’t say what those surveys were, who was being surveyed nor exactly what they were being asked nor when. It’s impossible to assess.

According to the writer of that Bloomberg piece, the “recent surveys” reference was actually to one survey and it was almost six months old (released on Jan. 17).

In this fast-moving space, not so sure that six months old constitutes “recent,” especially given that it looked at GoogleCheckout when it was just six months old itself, having been launched in June 2006. Technically, it’s not clear how old GoogleCheckout was at the time because it’s unclear when the surveys were conducted. All we know it that the results were published on Jan. 17.

That survey was conducted by JP Morgan Chase & Co., which is a well-respected source of such data. But the survey was based on interviews with 1,100 consumers, who may not readily know who to blame for a glitch. Therefore, failings that they attribute to GoogleCheckout might?in theory?be the fault of a site owner that wasn’t integrating it properly. Not saying that that happened, but there’s a lot of good reason to not put a lot of weight into a consumer survey when the topic is effectiveness of a highly-integrated checkout system.

Beyond the consumer perspective, how meaningful were the Checkout figures? Only six percent of the consumers interviewed had even seen GoogleCheckout, which isn’t surprising given its newness at the time. Theoretically, that’s 66 users out of 1,100. Of those users, 81 percent said they had “fair to poor” satisfaction levels with the product.

Problem one: with such a small sample size, it seems bizarre to merge “fair” and “poor.” How many of those 53-and-a-half interviewees said “fair” and how many said “poor”? Makes a big difference.

When the subject is the checkout system for an E-Commerce site and the interviewee is a consumer, one could make an argument that “fair” is respectable. How excited do you think a consumer is going to get about a checkout system? When rating the local brick-and-mortar, a picture-perfect flawless checkout would likely get a “satisfactory” or “fair” rating. A consumer would be quick to say “poor” or “very poor” with a bad checkout experience?ask Costco?but they expect competence and are not likely to give an especially high grade for it.

Now let’s take another look at that line from the Bloomberg piece: “In recent surveys, the world’s largest auctioneer found that less than one out of five users of Google’s Checkout online payment service was happy with it.” Setting aside that it was one survey and a not-so-recent survey at that, taking the leap from “81 percent said fair to poor” to “less than one out of five users were happy with it” is a stretch. It’s quite unclear whether those who said “fair” were not happy with it. For a checkout, “fair” might have been the selection of choice for those who were indeed happy.

And it may not be entirely fair to hold a year-old product to consumer criticisms?assuming they were indeed criticisms?when the product was much younger.

The survey also doesn’t seem to address that Google Checkout and PayPal are very different services, so the comparisons might be apple to watermelon.

GoogleCheckout has its issues but cautioning people against giving stats weight until the particulars are known is usually a good thing. If you’re going to address shortcomings, have the specifics to back it up.