What Drives Metrics?

Here’s my theory; see if it applies to your company. The marketing department realizes that it can’t answer questions about how well it is doing or whether it is making progress. So staffers review some of the solicitations they’re receiving from business intelligence software companies, view a few onsite presentations, and finally purchase a solution they think the company can afford. The data starts streaming in, and finally the marketing department has some measurements to use to determine how well it’s doing.

This is the bottom-up approach to managing marketing. By rights, a marketing department should use top-down planning. Christopher Escher, vice president of marketing for Responsys, explained the system the company recommends to its clients as follows:

  • Set goals that are in line with your business plan.
  • Determine what needs to be measured to ascertain whether the goals are being met.
  • Measure those things.
  • Receive reports of those measurements.
  • Modify plans to make sure that the goals are met if the initial reports show that the goals are in jeopardy.

Why So Much Data?

On the Web, most software originated with the ideas of technologists. This means that most reporting software gives users what technology can tell you rather than what you want to know. The difference is an order of magnitude. No one has time to read and digest what the technology can tell you, and, frankly, who cares what the technology can tell you? The genius of any business intelligence system is that it tells you what you need to know to optimize processes and improve results.

Back in July, I conducted a survey of marketing executives that showed that 85 percent of them don’t have access to adequate business intelligence (see “The Biggest Problem in Online Marketing”). Since then I’ve been writing — in both E-Commerce Guide and the EC Tech Advisor — about the problem of getting business intelligence from a strategic perspective (in E-Commerce Guide) and from a tactical perspective (in EC Tech Advisor — see “WebTrends Delivers Web Site Intelligence”).

What Marketing Intelligence Should Businesses Have?

The intelligence that businesses need to have depends on the strategic goals of the organization. Recently, I spoke about this topic with executives at three companies that seem to me to be doing a lot right: Angara, Responsys, and Unicast. Their answers reflected the fact that strategic marketing goals depend on the type of business.

Angara, which provides personalization via application service provider (ASP), including anonymous personalization to visitors who have never been to a site before, saw things from the perspective of the business-to-consumer (B2C) merchant or information site — Angara’s own clients’ perspective. Mark Metcalf, the company’s CEO, listed the classic measurements that retailers have the most trouble getting their arms around: “What are customer acquisition costs, all in, and what is the lifetime value of the member, all in? That’s what people should be trying to measure.”

But what constitutes those things? How do you define them?

As Metcalf pointed out, direct marketers have been measuring the cost of customer acquisition with great accuracy for many years. Metcalf continued, “For membership: How much do we bill them each year? What is [the] retention rate? What are [the] costs of retention we need to subtract (acquisition cost, ongoing cost to service, customer service costs, return rate, retention cost)?” Parallel values can be assigned for customers: How much do they purchase each year? What is the retention rate? What are the costs of retention that need to be subtracted?

The key to any metrics working for your business is that those metrics must permit you to assess your attainment of your strategic goals. This begs the question: What are your strategic goals? Let those goals drive your metrics collection, and you’re on your way down the road to having true business intelligence. Let meaningless metrics drive your strategy, and you’re on the path to obscurity.

Related reading

site search hp