Email Marketing ROI: How to Measure and Improve It
Most email ROI numbers are inflated by loose attribution and forgotten costs. Here is how to measure the real return, and the levers that move it, starting with a clean list.
Most email marketing ROI numbers are fiction. Not because people lie, but because they measure the wrong thing, attribute revenue too generously, and ignore the costs that quietly eat the return. The good news is that the real number is knowable, and once you can see it honestly, the levers that move it are obvious. This piece walks through both: a measurement method you can defend, and the handful of changes that actually shift the result.
The formula, and why most people get it wrong
The basic ROI formula is simple. The trouble is what you plug into it.
ROI = (revenue from email minus cost of email) divided by cost of email, times 100.
If you spend 1,000 and email drives 5,000 in attributed revenue, that is (5,000 minus 1,000) / 1,000 = 400 percent. Clean enough. The errors creep in on both sides of the equation.
- Revenue is over-counted. Last-click attribution hands email the full sale even when the customer was already going to buy. Counting a discount-code order as pure email revenue ignores the margin you gave away.
- Cost is under-counted. People remember the ESP bill and forget the rest: the hours spent writing and designing, the tooling, the cost of list acquisition, and the slow tax of a dirty list that drags every future send down.
Build an honest cost base first
Before you can trust the return, total up what email genuinely costs you each month. Be generous about including things, because hidden costs are how a 400 percent ROI turns into a real 90 percent.
| Cost line | Often forgotten? |
|---|---|
| ESP / platform subscription | No |
| People time (writing, design, QA, analysis) | Yes |
| List acquisition and lead-gen spend | Yes |
| Verification and list hygiene tooling | Yes |
| Discounts and margin given away in campaigns | Yes |
People time is usually the biggest line and the one nobody books. A single weekly newsletter can soak up a full day across drafting, design, and review. Price that at a real hourly rate and your cost base often doubles.
Attribute revenue conservatively
Pick an attribution model and stick to it, then resist the urge to claim every penny. Three workable approaches, from loosest to tightest:
- Last-click within a window. Credit email if the click happened inside, say, 7 days of the purchase. Simple, slightly generous.
- Incrementality testing. Hold out a random slice of your list, send them nothing, and compare their purchase rate to the group that got the campaign. The gap is the real lift. This is the most honest number you can get.
- Blended with margin. Whatever revenue you credit, convert it to gross profit before the ROI sum. Selling 5,000 at 40 percent margin is 2,000 of actual return, not 5,000.
If you only do one thing differently this year, run a holdout test once. The number that comes back is usually lower than your dashboard claimed, and far more useful.
The levers that actually move ROI
Once the number is honest, improving it comes down to a short list. They are roughly in order of impact for most senders.
1. Start with a clean list
This is the cheapest lever and the one people skip. Every dead address you mail costs you sending budget and, worse, drags your sender reputation down so that your good addresses land in spam. A list with 15 percent invalid addresses is not just wasting 15 percent of sends, it is suppressing inbox placement for the other 85 percent. Verifying before a big send, then keeping up regular email list hygiene, lifts the denominator and the numerator at the same time. If you have never cleaned a list, run it through a free email verifier and watch your bounce rate fall on the next send.
2. Cut the bounce rate
High bounces are a direct tax on ROI and a reputation signal to mailbox providers. Catch-all domains and stale role addresses are common culprits that basic checks miss. Our guide on how to reduce email bounce rate covers the mechanics. The short version: validate, segment out the risky addresses, and stop mailing people who have not opened in a year.
3. Improve deliverability
You cannot earn revenue from an email nobody sees. Authentication (SPF, DKIM, DMARC), consistent sending volume, and engaged recipients all push more mail to the inbox. Work through improve email deliverability if your open rates have quietly slid.
4. Segment and personalise
One message to everyone is the lowest-return way to send. Splitting by behaviour, purchase history, or lifecycle stage routinely doubles revenue per email without adding a single subscriber. This is where margin-aware targeting pays off: send the discount to the people who need it, the full-price launch to the people who do not.
5. Send less, but better
More sends can lower ROI by burning engagement and pushing unsubscribes. The metric to watch is revenue per email sent, not total revenue. If an extra weekly send adds 5 percent revenue but costs a day of work and a dip in deliverability, it is a net loss.
A worked example
Say you mail 50,000 contacts monthly. Platform cost 400, people time 1,600, so total cost 2,000. Last-click dashboard claims 18,000 in revenue, a tidy 800 percent ROI. Now apply the honest method: 40 percent margin brings it to 7,200 of gross profit, and a holdout test shows only 70 percent of that was incremental, so 5,040. Real ROI is (5,040 minus 2,000) / 2,000 = 152 percent. Still good, but a different conversation than 800 percent, and now you know cleaning the list (which cuts the 10,000 dead contacts you were paying to mail) drops cost and lifts placement at once.
FAQ
What is a good email marketing ROI?
Industry averages get quoted in the hundreds of percent, but those use generous attribution and ignore labour. A defensible, margin-and-incrementality-adjusted ROI of 100 to 300 percent is a healthy, real result for most senders. Be suspicious of your own four-figure numbers.
How often should I clean my list to protect ROI?
Verify before any large or important send, and run a full hygiene pass at least quarterly. Lists decay by roughly 20 to 30 percent a year as people change jobs and abandon inboxes, so a once-a-year clean is already behind. For the full picture see the complete guide to email verification.
Does cleaning a list really change revenue, or just costs?
Both. Removing dead addresses lowers your send cost directly, and it raises inbox placement for your valid contacts because mailbox providers stop seeing bounces and complaints from your domain. That second effect is the bigger one, and it is invisible until you measure inbox rate before and after.