Revenue per email (RPE) is one of the most valuable metrics (aside from ROI) that you can track, but few organizations, large or small, actually do. In this column, I’ll talk about the value of RPE, its strengths and weaknesses as a measure of success, how to calculate it, and how to use it effectively.
I’m a big fan of return on investment (ROI); it lets you know precisely how much profit you’re making from your email efforts. But ROI is often onerous to calculate, not because the formula is difficult, but because companies have a hard time getting their arms around true “costs.”
This is where RPE can be a good alternative. You don’t need to know costs. Unlike ROI, RPE doesn’t go directly to your bottom line – it doesn’t reflect profit (or loss). That’s a downside. But it does give you a feel for how much revenue you are generating from your email list, which is a great improvement over just having open and click-through rates.
RPE is simple to calculate:
Revenue generated/(Email quantity sent – number of bounces)
With my clients, I actually use what I call an RPME figure – it’s the revenue generated for each 1,000 email messages assumed delivered (email quantity sent – number of bounces). To calculate this, just multiply the RPE by 1,000. Or divide the assumed delivered figure (the denominator) by 1,000. RPME makes it easier to spot small variances when you’re doing comparisons. Just be sure you clarify that you’re using an RPME, not a straight RPE.
A recent survey by MarketingSherpa found that just 33 percent of email marketers track RPE; and no one reported tracking ROI! The only metric less tracked than RPE was social sharing rate (18 percent), a relatively “new” statistic. In the same survey, 90 percent or more of the respondents were looking at open and click-through rate; so why is RPE so underutilized?
Unlike open and click-through rates, RPE doesn’t magically appear in your email report. That’s because, in most cases, the tool you use to send email isn’t one that also tracks revenue from that email. You have to track revenue using your web analytics or other program, and then either pull it into your email reporting dashboard (some solutions allow you to set up an automatic pull on a regular basis) or a spreadsheet to get the calculation.
Adding to the difficulty in calculating RPE, some marketers still don’t have a reliable way to track revenue generated from email campaigns. There’s really no excuse for this in 2011. Most web analytics packages have a way to track revenue, including Google Analytics, which is free. If your sales happen offline, you’ll have to work up a separate process to track your phone, direct mail, or in-person revenue generated from either.
Bottom line: if you aren’t able to get reliable figures on sales generated from your email efforts, it’s time to make it a priority and do whatever it takes to get there.
You can (and should) look at a couple different RPE figures:
- RPE by individual email effort
- RPE by email campaign (multiple efforts)
- Overall monthly RPE
- Calculate your assumed delivered rate, which is the total quantity of email sent in a month divided by the total number of bounces
- Divide your total revenue for the month by your assumed delivered rate
- Monthly RPE by business unit, product, or other subset of your organization
- Quarterly and annual RPE
Once you begin calculating RPE, you have a new tool in your belt to use in evaluating and optimizing your email performance. Here’s a quick case study.
This was a direct response email for a large publishing company; my very first project for them. I analyzed the control and, using the metrics from previous sends, worked to come up with a test version that would generate more revenue.
When I first looked at the metrics from send (an A/B split between the control and my test version), my heart sank – how could the test have generated a 16 percent decrease in the click-through rate?
But then the revenue figures came in.
Although the test generated a few less orders (the conversion rate was 4 percent below the control), the orders were larger. We saw a 16 percent lift in the average units per sale and a 32 percent lift in the average order. Bottom line: the test had lifted RPE by 26 percent. Would you celebrate a 26 percent increase in revenue, even if it meant a 16 percent decrease in click-throughs? If it’s a direct response campaign, the answer has to be yes.
RPE involves a bit more work than open and click-through rate, but it’s a much better criterion to use to measure success. When you’re testing (which you should be), you might think that the version that generates more clicks is likely going to produce more revenue. But as you can see above, that’s not always the case.
Until next time,
Jeanne is off today. This column was originally published on July 11, 2011 on ClickZ.