Catalogers Offer Views On A Metric Too Many

When it comes to analytics, is there such a thing as too much information? Very much so, and here’s a nifty trick to find out if you’re going too far: “If you can’t remember all your metrics, you are probably measuring too many,” says Bruce Detweiler Breckbill, vice president of direct sales for Lehman’s, a Kidron, OH-based marketer that sells a variety of household products that don’t rely on electricity.

For Breckbill, metrics can be broken down in to three categories: Those for which accurate measurements aren’t available, those that are available but irrelevant, and those that are actually useful.

Take orders per customer service rep per hour, which for Breckbill falls somewhere between irrelevant and unavailable. Some of his reps focus on entering mail orders, as roughly 20% of Lehman’s orders come in via post, with a check or money order. And since there is a fair amount of explanation that electricity-free products requires, some of his best, most knowledgeable reps spend their time answering questions as opposed to pushing for an order.

Breckbill also doesn’t look too carefully at average order size. Yes, he does want his reps’ conversations to have a certain amount of cross- and upsell, but “I don’t want to push for that at the expense of good customer service,” he said. “I don’t know how to measure that without them being pushy. I want them to be polite.”

So what does he look at? Order volume and demand dollars by channel; early catalog responses (such as just after a mailing); early product sales (to determine whether certain items are moving more quickly than anticipated, requiring a restocking); and abandon rates for the Web site (a spike in these can indicate the site isn’t working).

There are hazards even in these, however. “Don’t get tied up in daily metrics – they can give you headaches really fast,” Breckbill cautioned. He prefers to look at his sales metrics on a year-over-year basis for a long-term view of his company’s progress – and on a weekly basis to identify current sales trends.

Another potential pitfall is using figures that are so high-level as to offer skewed pictures. For instance, order value by channel may reveal that the sales force channel, which serves industrial clients, brings in a series of high-value orders. But this information can be meaningless unless one also adds in the fact that the sales force channel brings in a very small percentage of revenue.

A marketing manager who deals with these figures on a day-to-day basis will probably know to factor in these sorts of anomalies. But marketer can run into trouble when presenting aggregated figures to higher ups, especially if commonly held knowledge, such as the difference between sales force sales volume and that of other channels, is not conveyed as well.

Then there are the metrics which are measurable, but irrelevant. Lehman’s sells items with a variety of size, weight and fragileness. A 75-pound iron fireplace grate and a glass lamp both present packing challenges, and both require time and effort to ensure they reach their buyers intact. Measuring the volume of items shipped would be meaningless – and might induce line workers to cut corners in the process.

For Catalog Music Corp., metrics allow the firm to evaluate new sources of prospects. The average life span of those buying its classic country music is around a year, making evaluating customer sources extremely important.

Martin Davis, the Nashville, TN-based firm’s managing director, spends most of his time looking at “weekly new to file” numbers, both in aggregate and by media channel. His concern is ensuring he has a consistent revenue stream.

Davis also tracks one-year value by original media. No, he’s not looking to extend the value of current customers (Catalog Music Corp. doesn’t even have a continuity program, although “we’d sure like to,” he said), but rather to determine where to place his marketing dollars. He also keeps alert for fulfillment problems by monitoring backorder rates on a daily basis.

Both of these marketers have worked to identify – and limit – the truly relevant metrics they track. As Al Bessin, a partner at Lenser, a direct marketing consultancy notes, “Metrics and measurements become self-serving in large corporations,” he said, recalling one company that tracked an unwieldy 205 metrics.

Metrics should be limited to a handful of essential functions, according to Bessin. They should enable organizations to allocate marketing resources; provide budget justification to higher-ups; measure customer experience; assist in projecting supply and operational requirements; help identify problems and errors; support stakeholder reporting (as is needed in financial documents and the like); and provide reliable indicators of progress regarding long-term business plans.

If a metric’s function can’t be classified within the list above, it doesn’t belong on a dashboard, Bessin said.

Bessin also warned against metrics stagnation. Just as businesses evolve over time, so should measurement reporting. Product mixes change, and may have an impact on customer lifecycles. Customers change behavior.

Even the way customers interact with a business may change – call centers, for instance, have moved from order-taking channels to informational channels. Any marketer who doesn’t understand how this is reflected in sales rates, and the impact these seemingly unprofitable call center contacts have on sales made through other mediums, will have a skewed view of the business.

He also cautions against incorporating irrelevant data into measurements. “Beware of industry averages,” he cautioned. “The industry is not you and your company. What’s most valuable is how you did versus a previous period of time.”

Other measurements that can lead marketers down blind alleys are averages. “Averages are horrible. Are your order values distributed along a bell curve?” Most marketers, he said, have price points that bunch in clusters.

If a marketer bases, for instance, a free-shipping offer on averages, upper-level customers, who are well above that order size, will cheerfully take advantage of that offer without providing any more value. Conversely, the lower-value customers won’t be close enough to that average to up their order size enough and take advantage of the free shipping.

Bessin advocates marrying a high-tech piece of equipment – a marketing dashboard – with a low-tech one – a calendar. “If you change the way you measure something, write it down on a calendar,” he advises. Doing so will allow marketers to understand why changes may occur in year-over-year trends. Additionally, marketers should note any unusual events that may have an impact on their business on their calendars.

Bessin cited Hurricane Katrina as one such event – sure, year-over year numbers from September 2005 may look terrific, but marketers need to remember factors beyond their control that influenced sales.

Finally, Bessin cautioned against becoming too obsessed with data updates. Unless a marketer is looking at data that will allow him to immediately identify and correct a problem, there is no need to micro-analyze returns. “CEOs like to look at hourly reports,” he said. “Drug companies love those CEOs.”

Bessin, Davis and Breckbill shared their wisdom during a session at the Direct Marketing Association’s annual conference in San Francisco.