How to Measure What Matters

LinkedIn Article by Tim Williams 
November 10, 2013

As professional service firms like advertising agencies begin to experiment more with the concept of outcome-based compensation agreements, it’s essential to understand how to measure what matters.

Most measures of performance––sales, market share, stock price, etc.––are lagging indicators that only tell us where we have been. What we need are leading indicators. We need to measure not just brand success, but understand what are the precursors of brand success.

For example, your personal credit score contains five weighted measures: payment history, amounts owed, length of credit history, new credit and types of credit used. Why just these five among the hundreds that could be measured? Because these five have been empirically tested for predictive value. In short, they are the leading indicators that best predict a person’s credit worthiness.

Ask most marketers how they measure success and they’ll likely respond by saying they focus mostly on such performance measures as sales volume, sales growth, market penetration, share of wallet, etc. But that’s like looking in the rear view mirror to figure out where you’re going. These types of lagging indicators only report on what has been. They have zero predictive value.

 

Lagging indicators are simply a result. Leading indicators are a measurement that explains or predicts the result. Most marketers spend their time and energy analyzing the result instead of observing and understanding the process that leads to the result.

Identifying drivers of success

Most leading indicators never appear on a financial statement or sales report, but they have tremendous predictive power ––that is, they will drive the numbers that ultimately appear on the financial statements. The correct leading indicators will lagging indicators.

Working with Northwestern University, a social media analysis company called MotiveQuest developed a metric called the “Online Promoter Score,” a measure of the frequency and willingness of consumers to advocate strongly for and recommend a brand. MotiveQuest worked with the car company MINI and their advertising agency Butler Shine Stern & Partners to measure the relationship of the Online Promoter Score to sales, concluding that this leading indicator has strong predictive power.

Similarly, a major food company found that “number of recipes distributed” is a reliable predictor of sales. A large automobile manufacturer has found they can reliably correlate sales with the number of test drives. The marketing goal then becomes to focus on the leading indicators, not just a nebulous directive to increase sales.

Looking ahead instead of behind

This is not to say there isn’t a place for lagging indicators in analyzing the success of marketing efforts. Some lagging indicators – such as incremental profits generated from a marketing campaign – are important and relevant measures of marketing success. The same is true with lagging indicators like brand penetration and average price per unit.

But many traditional measures of success are the result of historical practices rather than a careful study of cause and effect. Correlation is not the same thing as causation. For example, while sales are the most common “hard” metric of success, recent research shows that campaigns that focus on reducing price sensitivity are more effective than those that focus on volume. In other words, value is more important than volume, and value share more important than volume share.

The challenge – and the opportunity – is to engage marketing teams in identifying the metrics that predict a brand’s success, and then work to improve them. Here are only a few examples:

EXTERNAL INDICATORS

  • Customer compliments (Number of compliments from customers in a given period)

  • Customer complaints (Number of complaints from customers in a given period)

  • Customer referrals (Number of referrals received from customers in a given period)

  • Customer suggestions (Number of suggestions from customers to improve the brand)

  • Online endorsements (Number of endorsements of brand published online in a given period)

  • Positive brand experience (Degree to which customers have had positive experience with brand)

  • Brand knowledge (Percent of target population knowledgeable about brand)

  • Brand fame (Degree to which brand is seen as having momentum)

  • Brand likeability (Degree to which brand is seen as likeable or “like me”)

  • Positive press coverage (Degree to which news about the brand is positive)

  • Willingness to search (Percent of target that will delay purchase if our brand is not available)

  • Willingness to recommend (Percent of customers that would recommend brand to a friend)

  • Brand quality ratings (Rating of brand’s quality compared with others in category)

  • Brand functionality ratings (Customer ratings of brand performance)

  • Brand value for price (Rating of agreement that brand represents good value for the money)

  • Customer satisfaction (Percent of customers who give high ratings to their experience with brand)

  • Service satisfaction (Degree to which customers are satisfied with service from company)

  • Ideas from channel partners (Number of ideas received by company from channel partners in the distribution chain)

  • Innovation pipeline (Number of new products or services in development)

INTERNAL INDICATORS

  • Positive employee brand attitudes (Degree to which employees have positive views of the company and the brand)

  • Belief company is living the brand (Degree to which employees believe the company practices are in line with the brand’s stated values)

  • Employee satisfaction (Degree to which employees are satisfied working for the company)

  • Likelihood of staying with company (Intent of employees to retain their employment status)

  • Would refer company to friend (Likelihood employees would recommend the company to a friend)

Knowing the metrics that matter should be part of the intellectual capital marketers and their agency partners bring to the party. Not all measures are created equal.

Comment