Discerning Distinctions in Buying Behavior

OK, I know that many of you look to justify your Web presence based primarily on the number of new customers acquired there. Though I do not believe that this is the only, or even the most important, measure of success for most businesses (refer back to the last several weeks of this column if you want to know why I feel this way), I certainly accept that it is one metric that matters, and in some rare instances it may in fact be the only metric that matters.

It is often these customer-acquisition-focused businesses that aim for high click rates in their marketing, assuming that a click is a necessary first step to a customer’s starting down the path to purchase. But anyone who has been marketing online for more than a day or two already knows that all clicks are not created equal. So how do we use data analytics to determine which customers are worth encouraging, which tire-kickers are likely to become paying customers?

Watch them shop. Yes, I mean use your tracking capabilities to follow that shopping cart around and watch for the behaviors that distinguish the surfers and browsers from those who complete a transaction. Then track those distinctions back to the source of the initial lead to look for patterns. Though many marketers source a successful transaction back to the site it linked in from, few take the necessary next step to source the creative message that generated the link. Those who do often find that appropriately targeted messaging has a lot to do with which new arrivals become customers.

Perhaps the more important application for analytics here, however, is a focus on increasing purchase likelihood rather than just on understanding those who purchase. Smart retailers look for displays that cause shoppers to lose interest; online sellers should do the same. When you see a correlation between cart abandonment and a particular point in your site, experiment with changes to the site before assuming that the customer is necessarily the problem.

Great merchandisers constantly test every tiny aspect of the customer experience — not to look for right and wrong answers (what can be so wrong about someone’s unique experience?), but simply to alter things that do not seem to be producing the desired result.

Because redesigning a Web site can feel more complicated than changing the merchandise displayed on the row-seven end-round, we see less of this sort of testing and learning going on in the e-commerce arena. But it ought to be happening more. After all, we have less history in this shopping world, and customer experience is still forming and evolving with regard to shopping online, whereas behaviors and expectations change less quickly in the more established world of physical-space retailing.

If you set up your Web analytics to track behavior, you’ll optimize that effort by also setting up your site to easily enable revision, then you’ll test for changes in results.

Sounds simple, but in fact it takes a lot to do this well — sometimes entire organizations have to be trained to think differently about the role of Web pages. But for those who do it well, it is extremely worth it.

Profitability comes from knowing who to focus your marketing message on, learning how to improve the buying experience of those you do attract, and increasing their propensity to complete the sale (as well as to increase the size or frequency of sales).

Now it’s getting interesting, eh?

Related reading

tencent_emily-ma_featured-image
kenneth_ning_emarsys_featured-image
bounce-370x229
site search hp
<