How to Measure Email Marketing ROI

A formula for email ROI that uses revenue and cost you can actually name, plus what to exclude so the number is not theatre.

A workable definition: email ROI equals attributed revenue minus cost, divided by cost. Cost includes the tool, the data you paid for, and the time to write and approve. Revenue includes only money you can connect to a reply, a tracked link, or a deal the team agrees came from that send. It does not include a multiple you found in a survey.

Worked shape, with symbols rather than a fake company result. Revenue R, cost C. ROI = (R − C) / C. If R is 0 because nobody will own the attribution, the honest statement is that ROI is unknown, not that email 'returns 36 to 1'.

What to count

For a newsletter, you may only be able to count assisted conversations. Say assisted. For a cold programme, count customers and meetings, and put the reply on the same CRM record as the send so the argument is inspectable.

Sendblox keeps that send and that reply together. It does not produce a magic ROI. The formula still needs someone to decide which revenue belongs to the mail.

Before you publish a change for email marketing ROI, write down the date, the record you edited, and the one message you used as a test. A week later, check whether that test still matches production. Provider rules, DNS includes, and CRM fields all move. A page that was true in general can be wrong on your domain if nobody looked again. Keep the send small until that check is dull. One contact, one objective, a person reading the draft. If the test fails, stop the campaign. Do not widen it to see whether a larger list forgives a broken setup.

Before you publish a change for email marketing ROI, write down the date, the record you edited, and the one message you used as a test. A week later, check whether that test still matches production. Provider rules, DNS includes, and CRM fields all move. A page that was true in general can be wrong on your domain if nobody looked again. Keep the send small until that check is dull. One contact, one objective, a person reading the draft. If the test fails, stop the campaign. Do not widen it to see whether a larger list forgives a broken setup.