Email marketing ROI: why $36 to $1 is not your number, and what it actually costs to run

The $36 back for every $1 spent on email is a figure Litmus cites as an average across the marketers it surveys (Litmus). It is not a forecast for your business. Litmus’s 2025 State of Email survey covered nearly 500 marketing professionals. In it, 35 percent of marketing leaders reported $10 to $36 back per dollar and 30 percent reported $36 to $50. Another 21 percent said they do not measure email return at all (Litmus).

What it costs is easier to pin down. At a 5,000-contact list the sending platform runs from about $9 to $100 a month. The work on top is your own hours, a freelancer by the hour, or an agency retainer that typically starts at $1,000 to $3,000 a month. Here is where each number comes from, and how to work out your own in about five minutes.

The short version

• Software at 5,000 contacts costs about $75 a month on Mailchimp Essentials and about $100 on Mailchimp Standard or Klaviyo. Brevo starts at $9 a month because it bills by emails sent, and that entry plan covers 5,000 emails a month.
• An agency retainer for a smaller business typically starts at $1,000 to $3,000 a month on top of the software, according to InboxArmy’s published pricing guide.
• The $36 to $1 return is self-reported. In Litmus’s 2025 survey, 21 percent of marketing leaders said they do not measure email return at all.
• Open rate is no longer a dependable performance measure. Apple Mail Privacy Protection downloads remote content in the background whether or not the person reads the email.
• Google tells bulk senders to keep the user-reported spam rate below 0.1 percent and never let it reach 0.3 percent.
• Your own number is email-attributed revenue minus total email cost, divided by total email cost, measured over ninety days.

What does email marketing cost at 5,000 contacts?

There are two bills, and a pricing page shows only one of them. The first is the sending platform, at about $9 to $100 a month for a 5,000-contact list. The second is the person doing the work.

Route	Monthly cost at 5,000 contacts	What it covers	What is still on you
Platform only, run in-house	About $9 to $100	Sending, templates, automation, reporting	Strategy, copy, design, segmentation, deliverability, every send
Platform plus a freelancer	Platform cost plus hourly fees, from about $15 to $40 an hour on freelance marketplaces	Campaigns built, written and sent	Strategy, approvals, cover when they are unavailable
Platform plus an agency retainer	Platform cost plus about $1,000 to $3,000	Strategy, calendar, copy, design, segmentation, automation, reporting	Approving work, supplying offers, answering the replies

Read the right-hand column before the middle one. The platform tier is the one figure that is fixed and published, and it is the smallest number in the table.

The platform prices below were checked on 2 October 2026 against published breakdowns, because the vendors’ own pricing pages change with region and promotions.

Platform	Price at about 5,000 contacts	How it bills	Source
Mailchimp Essentials	$75 a month	By contacts	EmailVendorSelection
Mailchimp Standard	$100 a month	By contacts	EmailVendorSelection
Klaviyo Email	$100 a month	By active profiles	EmailToolTester
Brevo Starter	From $9 a month for 5,000 emails	By emails sent	EmailToolTester

Two details matter more than the headline rates. Brevo’s $9 buys 5,000 emails a month, which is one send to a 5,000-contact list. Send weekly and you are on a higher tier. And since 18 February 2025, Klaviyo requires customers to be on an email plan that matches or exceeds their active profile count. It moves an account up a tier automatically at the next billing cycle when the count goes over (Klaviyo Help Center). A large list you rarely mail still sets the bill.

On the labour side, InboxArmy’s pricing guide says retainers “typically start at around $1,000 to $3,000 per month for smaller businesses”, and run from $3,000 to $10,000 or more for mid-sized and large ones. It puts freelancers on the marketplaces at $15 to $40 an hour, rising to $250 an hour or more for specialists (InboxArmy). InboxArmy is an email agency, so read these as asking prices and not as an independent survey. The same three-way choice runs through every channel, and the full cost comparison of an agency, a freelancer and an in-house hire puts the salary arithmetic next to the retainer.

Photo: olia danilevich on Pexels.

Where does the $36 to $1 number come from?

It comes from Litmus, which sells email testing software and surveys email marketers about their programmes. Litmus cites an average return of 36 to 1 (Litmus). Its page on email return gives the spread behind that kind of figure, from its 2025 State of Email survey of nearly 500 marketing professionals worldwide (Litmus).

What marketing leaders told Litmus in 2025	Share
Receive $10 to $36 for every $1 spent on email	35%
Receive $36 to $50 for every $1	30%
Receive more than $50 for every $1	5%
Do not measure email return	21%

That is a real survey of working marketers. It is also people reporting their own results, and three things follow from that.

First, the figures are what respondents said, not what an auditor counted. The same page says the most common way respondents measure return is by analysing engagement metrics, at 59 percent, ahead of direct revenue attribution at 41 percent.

Second, one in five do not measure return at all. Litmus notes that share was 36 percent in 2023.

Third, the survey reaches people who work in email and chose to answer questions about it. A business where email did not work is less likely to be in the sample.

Litmus sells email software, and the figure is widely repeated by companies that sell email platforms. None of that is hidden, and none of it makes the research dishonest. Read it the way you would read any supplier’s figures about its own category.

How do I work out my own email return?

The formula is email-attributed revenue minus total email cost, divided by total email cost. Run it over ninety days, because one large send distorts a single month.

1. Pull ninety days of email-attributed revenue from your sending platform and from your analytics separately. They will disagree. Use the lower of the two.
2. Add up every email cost in the same ninety days. Subscription, agency or freelancer fees, design, deliverability tools, and your own time at a rate you would pay someone else.
3. Subtract and divide. The answer is what one dollar of email spend returned. Express it as a ratio so you can set it beside the published figures.
4. Compare it with the other use of that budget. For a small business that is often paid search or paid social.

Here is a worked example with round numbers, chosen for the arithmetic and not taken from any account. A business pays $100 a month for its platform and $1,500 a month to an agency, so ninety days costs $4,800. Its platform reports $18,000 of email-attributed revenue and analytics reports $14,400, so it uses the lower figure. $14,400 minus $4,800 is $9,600. Divided by $4,800, that is a return of 2.0, or three dollars of revenue for every dollar spent. That is a good result, and it is a long way from 36.

The weak point is the measurement, not the arithmetic. A sending platform credits email with a sale made inside a set window after an open or a click, and some of those sales would have happened anyway. Two tools looking at the same week will give two answers. It is the same problem described in why Meta and GA4 report different conversion numbers: different attribution windows and different definitions of a conversion.

The cleaner test is a holdout. Withhold campaigns from a random ten percent of the list for a quarter, then compare revenue per contact with the group that received them. The gap is what email caused.

The email marketing service page sets out how that tracking gets built before the calendar does.

What is a good email open rate in 2026?

There is no single dependable figure, and a page that offers one without naming its dataset is guessing. Two large published sets show why.

Benchmark	Average open rate	Average click rate	What was measured
Mailchimp	35.63%	2.62%	Mailchimp campaigns sent to at least 1,000 subscribers, page last updated December 2023
GetResponse	39.64%	3.25%	More than 4.4 billion messages sent by GetResponse customers in 2023, active senders with at least 500 contacts

Each set is one vendor measuring its own customers, which is a large sample of that vendor’s customers and not a sample of all email.

Opens themselves are the bigger problem. Apple says Mail Privacy Protection “downloads remote content in the background by default”, whether or not you engage with the email, and stops senders learning when and how many times you opened it (Apple). An open can be recorded for a message nobody read. How much that inflates your figure depends on how many of your subscribers use Apple Mail.

Track these instead:

• Click rate, which needs a deliberate action.
• Click-to-open rate, which tells you whether the message matched the subject line.
• Conversions and revenue per recipient, the figures the business runs on.
• Unsubscribe rate, a plain measure of whether people want the emails.
• Spam complaint rate, the one that can stop your email arriving.

Which deliverability rule decides whether any of it works?

Gmail’s spam-rate limit. Google’s sender guidelines have applied since February 2024. Anyone sending close to 5,000 messages or more a day to personal Gmail accounts has to authenticate with SPF and DKIM and publish a DMARC record. They also have to support one-click unsubscribe on marketing messages (Google Workspace Admin Help).

Google tells senders to keep the user-reported spam rate below 0.1 percent and to prevent it ever reaching 0.3 percent. Since June 2024, bulk senders above 0.3 percent are ineligible for mitigation, and become eligible again only after the rate stays below 0.3 percent for seven consecutive days (Gmail sender guidelines FAQ).

Read that as a budget limit. Three people in a thousand pressing the spam button is the whole allowance. On a send to 5,000 Gmail users that is fifteen people, and one badly targeted send to a cold segment can produce that in an afternoon. A retainer buys very little while Gmail is filtering the sends.

If any part of your list sits in the European Union or the United Kingdom, consent is the other limit. The rules on how you collect and record it overlap with the consent requirements that now apply to small businesses in the EU and UK. A list built without recorded consent is a risk you pay a subscription to keep.

Photo: Kampus Production on Pexels.

When is email marketing not worth paying for?

When there is no audience for it to work on, or no reason for that audience to buy again. Email mostly sells to people who already know you. These are the five situations where we would advise against a retainer:

• Your engaged list is small. Our rule of thumb is under about 500 people. Build the list first. A $1,500 retainer against 300 contacts is hard to earn back.
• The customer buys once and never again, and there is no referral or review request to run in its place.
• The list was bought, scraped or imported from an old system with no record of consent. That is a spam-complaint problem waiting for a send date.
• There is nothing to say every month. Sending often with little to say is how complaint rates climb.
• The site does not convert the traffic it already has. Sending more people to a page that does not work multiplies the problem. Conversion optimisation is the cheaper fix.

Where email does earn a retainer, in our experience, is the automated sequences more than the newsletter. Welcome, abandoned cart, post-purchase, win-back and replenishment are built once and then run unattended. That is why we start with email automation before a monthly campaign calendar. Choosing between funding this and funding acquisition is the same decision covered in SEO versus paid search, and it turns on the same question: how soon do you need the revenue.

If you want the measurement built before the calendar starts, the email marketing work begins there.

The bottom line

A published average tells you what other people said about their own email. The number to act on is the one your last ninety days produced, checked against a group that did not get the emails. Work that out before you compare platforms or retainers.

Questions people ask about this