'EmailGeekonomics': Time For Some New Email Metrics

How is email marketing like baseball? Both use a set of generally accepted statistics to measure player (campaign) performance and how it contributes to the team's (company's) success in meetings its goals (winning games/generating revenue).

However, baseball organizations and other major-league sports may be farther up the evolutionary ladder than many email marketing organizations, because they are using new metrics that provide better insights and more accurately measure and predict player performance.

Baseball Invents More Meaningful Statistics

I've been fascinated with sports statistics since I was a youth, opening the San Francisco Chronicle's daily Sporting Green section to see how Willie Mays ranked in Major League Baseball's various categories.

Today, teams and statisticians combine basic stats like hits, walks, runs scored, runs batted in, and more into new, more involved formulas that more precisely measure a player's individual performance and his actual value to the team's overall performance.



Why the OBP Beats the Batting Average

Like the open rate in email, the batting average is the classic measure of batter performance, but it doesn't adequately capture a batter's value. For example, power hitters often get walked intentionally. Walks, which are not reflected in batting average, often lead to runs scored.

A better metric is the on-base percentage (OBP), which measures all the ways a player can get on base, including hits, walks and hits by the pitcher (not including errors, interference or fielder's choice). These could advance another base runner or score a run. This more accurately represents the player's at-bat performance and his contribution to the game.

This new world of baseball statistics even has its own case study. "Moneyball: The Art of Winning an Unfair Game," by Michael M. Lewis (2003; W.W. Norton & Co., Inc.) documents how the Oakland Athletics used more empirical measurements of player performance to build a winning team with less money. (OK, the verdict may be out on the A's success.)

Time to Change the Email Stats Roster

More email professionals are realizing that many standard email metrics are flawed or misused as yardsticks of email performance. (See an explanation here: "Are You Using The Right Metrics?").

These lesser-known metrics could foster some discussion to develop more comprehensive and meaningful email metrics:


  • ROEM (Return on Email Volume): Certain email programs, such as  triggered emails, deliver much greater return or revenue than others based on volume sent. This analysis shows management that adding that recurring birthday email program may actually provide better ROI than simply sending another broadcast email.


  • ARED (Average Revenue Per Email Delivered): Used by a fair number of ecommerce companies, ARED provides a simple way to measure the pure effectiveness of your emails in generating revenue. Calculated by dividing total revenue by number of emails delivered (not just sent).


  • Percent Active: Measures how many subscribers (expressed as percentage of email list) have opened or clicked over a specific time period, such as six or 12 months.


  • List Hurdle Rate: How many new subscribers you must attract not just to replace those lost to churn -- unsubscribes, spam complaints and inactivity -- but also to meet your annual (or other time frame) list-growth goal.
  • Churn to Clicks Ratio: The ratio of unsubscribes and spam complaints per message to unique clicks. Benchmarked over many messages, this comparison allows you to understand the correlation between your positive call to action content and links and negative churn.

Cost Savings

  • Contact Deferral Rate: This traditional customer-service metric measures which alternative communication channels customers use instead of the call center (online communities, email, direct mail, FAQs, social media, etc.). With email, emails can be designed to educate and solve FAQs for customers, greatly reducing costly call center calls.

Reach/Brand Lift/Viral Effect

  • SRV (Subscriber Referral Value): Measures subscriber value in terms of additional revenue generated from existing subscriber referrals via share-to-social and forward-to-a-friend links.
  • Viral Reach: Calculation of a campaign's reach through subscriber postings in social networks, discussion groups, forwarded email and other online sources.

What Do You Want to Measure?

Like an ocean liner, the email industry can't reverse course instantly. However, groups like the Email Experience Council's Measurement Accuracy Roundtable is helping to steer the ship in the right direction.

The group will soon announce new standards and definitions for a number of core email metrics. As part of this standardization process, perhaps we can also create a new set of business-oriented metrics that will generate results that provide greater context for the C-suite as well as the marketing department.

Do you use a lesser-known email metric? Please share it and its value to your organization in the comments section below.

Until next time, take it up a notch!

4 comments about "'EmailGeekonomics': Time For Some New Email Metrics".
Check to receive email when comments are posted.
  1. Bruce May from Bizperity, May 20, 2010 at 3:45 p.m.

    As a results focused executive, I like ARED (Average Revenue Per Email Delivered). That gives me an instant comparison between campaigns. Yet, perhaps the most valuable metrics here are those that reveal the reach and viral effect of social media. Perhaps in a future article you can talk about how you actually measure these. It's not easy to wrap your arms around metrics that exist on the other side of the social media divide.

  2. Loren McDonald from IBM Marketing Cloud, May 20, 2010 at 7:35 p.m.

    Thanks Bruce, I agree ARED is a good solid metric that basically tells you how effective are out emails at producing revenue. It cuts through the open, click, conversion, etc., etc. noise and gives you the bottom line. As such, it is great for benchmarking message to message, types of messages, etc.

    As for the social/viral metrics - that's a great idea, I'll plan to discuss that in a future column.


  3. John Caldwell from, May 21, 2010 at 1:41 a.m.

    Another thought-provoking post, Loren!

    I think what you're talking about here isn't so much a new set of metrics as much as advanced applications of existing metrics. At least for Revenue, Engagement, and Churn, as those are all based on a standard metrics.

    DISCLAIMER: "standard" metrics are NOT the same as "standardized" metrics (just like "delivered" is NOT the same as "deliverability"; but I digress....)

    Something to add to the deferral rate is any reduction in charge-backs due to email messaging. If one really wanted to get their geek on, they could drill that down to impact on payment processing fees.

    Reach/Lift/Effect should still compare to the basic measurement data to make any projections or estimate value.

    I totally like where you're going with this, but think it's focused more on applications of existing metrics than the creation of new ones.

    BTW - Props for pulling old-school terms like "Viral"! ;)

  4. John Getze, May 25, 2010 at 2:53 p.m.

    Great article Loren!!!!

Next story loading loading..