Measuring Sticks

The spoils of the holiday e-commerce season will be distributed unevenly. For every highflier like Yahoo, there will be a flop like Value America. But how can you tell which sites are headed in the right direction?

The uncertain outcome of high-priced Web site investment points to the need for customer-centric performance measures that serve as milestones on the journey to profitability. The tricky part is deciding what data points are most effective.

Technology companies have always had an affinity for measuring relative performance. Hardware makers have long boasted about such exotica as FLOPS, MIPS and seek times. And as spending on Web development has grown, a variety of new metrics has emerged. Alas, measuring the efficiency of the infrastructure is a fairly limited way to judge a Web site - you need to consider not only software and hardware performance, but also customer experience and behavior.

Companies like Service Metrics argue that there is a direct correlation between inadequate infrastructure and customer abandonment. But the new metrics make stronger linkages among capital investment, marketing expenditures and customer capture - and they're more relevant to measuring return on investment. Here's a look at some worth watching. Conversion Rate to Purchase. This is a measure of what percentage of total visits leads to purchases. If 100 people visit in a given span and five of them buy, the site has a 5 percent conversion rate. This metric can further be reduced to first-time or repeat-visitor conversions.

That's an important distinction, since there is evidence that the longer a site exists, the higher the conversion rate - suggesting that folks will visit several times before buying. This metric gives a holistic look at a site's overall effectiveness to drive purchasing, although it does not explain why someone bought. For that, you need to look at ...

Conversion Rate to Leads. This examines how sales are driven from various sales and communications channels: banner, broadcast, print. It can also be parsed by media strategy: advertising, promotion, public relations.

The Digital Sales Completion Rate. This provides the percentage of all sales confined to the Web, as opposed to those beginning at the site but finishing elsewhere, perhaps in a store or through an 800 number. If the goal is for your site to be an autonomous place of exchange, how close are you? Poor interface design, confusing instructions and slow downloads simply heighten costs as the toll-free line becomes the necessary transaction tool.

Average Cost of Customer Acquisition. Out of all the ways to get paying customers to a site, what is the average cost of capturing them? This metric can be an effective way to engage a deeper market analysis. If the average cost is going up, can that be attributed to higher capital or labor investment in the effort? Or is it due to higher marketing costs because of competition?

Brick-and-Mortar to Web Sales Ratio. This measure addresses an interesting question: If a company has a fairly fixed customer base, is a greater percentage of the customer base now buying via the Web over the historical channels? Even a rough measure would demand a high amount of discipline to collect very detailed customer information from all sales channels, on all occasions, then parsing it in a well-designed data warehouse. In industries where ownership of the customer relationship is becoming hotly contested, however, attempting some kind of channel cannibalization estimate might be unavoidable. These metrics share not only a vulnerability to revision as we continue to learn about the nature of the Web, but also a dependency on a larger benchmark from which to compare relative results. While any measure of internal performance over time is useful, we are better served in knowing not only that our conversion rate has increased 30 percent from last year but that the industry average conversion rate has increased 80 percent, consigning us to the bottom in our industry.

As metrics proliferate, it will be crucial that industry organizations become trusted, impartial brokers of industry data so members can compare internal against aggregate results. Although many e-commerce sites defy categorization, this is where the real value of metrics kicks in. Many associations, operating like government agencies, do not have the wherewithal or vision for such an undertaking. Dues-paying members should get them some; benchmarking accurate performance of e-commerce initiatives will depend on it.

- John Berry is an IT consultant in Bend, Ore.

Join the newsletter!


Sign up to gain exclusive access to email subscriptions, event invitations, competitions, giveaways, and much more.

Membership is free, and your security and privacy remain protected. View our privacy policy before signing up.

Error: Please check your email address.

More about Service MetricsYahoo

Show Comments