How to Use Key Performance Indicators

Many firms have metrics that they show to the outside world. Then there are the ones they use internally to monitor their progress.

These are called Key Performance Indicators (KPIs). They are selected measurements that together can provide a holistic view of the business, according to Rajan Maskay, senior vice president, engagement manager with KnowledgeBase Marketing Inc.

Let’s say your goal is to acquire new customers. Meaningful measurements would be:

*New customers added in month.

*Increase over last year in new customers per quarter.

*Increase over last year in new customer per quarter in a specific demographic group.

*Increase over last year in new customers per quarter in a holiday period.

Or, you might want to increase repeat purchases. You would measure:

*Frequency of purchases in one year.

*Ratio (frequency of purchases to opportunity to purchase) in one year.

*Moving average of customer purchases over six-month period.

KPIs can also cover revenue, margin, growth, acquisition, retention, churn, ROI, relationship strength and service time. And they can be assigned to brand measurements that may not be linked to transactions.

How do you determine which are the right KPIs for you? First you must determine your marketing objective, and make sure that it is:

*Aligned with your business strategy.

“*Derived from organization goals.

*Driving your marketing plans.

*Measured by KPIs.

The next step is setting the metrics. This requires breaking your objective into measurable outcomes, and then assigning KPIs to them (and they must be your KPIs, not the industry’s).

But the measurement must be done consistently, and with “meaningful frequency,” Maskay says. “You can’t just do things once a year.”

And it must be based on accurate data, delivered in a consistent way. “If it is input badly or changed midstream, the whole exercise is kind of moot,” Maskay says.

Occasionally, you may need to fix misaligned KPIs and the business objectives that drive them.

And don’t be daunted if you end up with ratios rather than actual numbers. “There’s a lot of information condensed in them, and you can use it as a roadmap,” Maskay adds.

Next comes the fun part: Human intervention.

“KPIs are not prescriptions for actions,” Maskay says. “If (direct mail) acquisition rates go down, you can’t just say we’re going to increase our penetration of the list. Analysis is necessary.”

He adds: “Computer systems that generate numbers are relatively dumb. They do what you tell them to do.”

Maskay offers these guidelines:

*A single KPI shows need for attention.

*A cluster of KPI’s provides better view—but not prescriptions for action.

The next step is opportunity assessment. Maskay notes that:

*KPIs are not self-sufficient.

*Assessment leads to recommendations for modifications to current strategy and plans (pricing/promotion/product, etc..

*This analysis may lead you to new opportunities—i.e., target universe, new product development, new channels.

Maskay also offers tips on how to analyze (a delicate issue when multiple departments are involved). First, there should be no finger pointing. (He concedes: “This is kind of hard to do.”)

Instead of assigning blame, team members should “hypothesize causes, and talk about why this may be happening.”

Aren’t there any stumbling blocks?

There are several. For one thing, your measurements may be misaligned with the business strategy. Or, you may have too many metrics.

“It’s easy to get into the death by analysis syndrome,” Maskay says.

Another problem occurs when you’ve got metrics without supporting human capital. Finally, you can’t just push this from the top down.

“All of us as worker bees have to accept that KPIs make sense,” Maskay says. “They’re like laws. You don’t agree with them? Try to change them, but for the time being, they are what the organization is working towards.”

How do you design KPIs? Here’s a checklist for what to do.

*Make sure they support your objective.

*Articulate what should be measured.

*Metrics must be internalized to organization.

*Define data required.

*Identify data sources, whether marketing or operational.

*Identify new data required.

*Plan measurement and reporting process or tool.

This article is based on a session at the Direct Marketing Association’s DMDNY conference.