Is email marketing still worth it in 2026?

Email marketing can be worth the investment in 2026. But “worth it” needs a budget, an audience and a result: which customers will you contact, what will the programme cost, and what additional value must it produce?

For an ecommerce business, start with the contribution from extra orders after discounts and variable costs. Then compare that with the full cost of doing the work.

Is email marketing still worth it? It can be, when the additional order contribution it produces exceeds the programme’s cost. Count software, labour and relevant setup or acquisition costs. Use industry ROI figures as context, then work out your own break-even point. Revenue credited to email is not automatically revenue caused by email.

What does the evidence say about email’s value?

There is evidence that emailed offers can generate additional sales, but it does not establish a universal return for today’s businesses. A study in Management Science, published online in 2016 and in print in 2017, analysed 70 randomised experiments at one large online ticket resale platform. Customers sent discount offers spent more during the promotion window than customers randomly not sent those offers. That is historical sales evidence in a particular setting, not a 2026 profit or retention benchmark.

Current industry research answers a different question. Litmus’s 2026 State of Email report says it surveyed 500 marketing professionals across the US, UK, Australia and New Zealand, asking about their company’s email ROI. Reported returns describe respondents’ programmes; they are not a forecast for yours.

Even your own attributed revenue needs interpretation. Klaviyo documents configurable attribution windows and a default last-touch model. Those rules assign credit to qualifying interactions. They do not establish which purchases would have disappeared without the email.

The useful question is therefore narrower than whether the channel “works”: can your proposed programme produce enough additional contribution to pay for itself?

What should you count as the cost of email marketing?

Count the resources required for the programme you are evaluating: platform and tools, production labour including owner time, setup and maintenance, and acquisition spending allocated to building the audience. A software-only budget leaves out the work of preparing, building, checking and maintaining the emails.

Use this suggested cost inventory:

Assign a value to owner time even when it creates no new cash payment. That makes this an economic-effort budget; keep a separate cash view if you also need to know what leaves the bank account. For a production-time log, use the guide to whether AI saves time on email marketing.

Match the budget to the decision. Starting a programme means evaluating its full cost. Adding a campaign to an existing programme means evaluating the extra work and spend that campaign requires. Avoid charging the same agency work or acquisition expense twice, and state how you allocate one-off setup across the period being assessed.

How many extra orders would make email worth it?

Divide the programme cost by the contribution from an additional order, provided that contribution is positive. This adapts the Small Business Administration’s break-even formula to an email-budget decision. It is a planning worksheet, not a forecast of sales.

Additional orders needed = programme cost ÷ contribution per additional order.

Here, contribution means collected order revenue after discounts and refunds, minus variable product, payment, fulfilment and shipping costs borne by the business. Keep the email programme’s cost separate so it is not deducted twice. If product margins vary, use separate calculations or a justified expected order mix.

Consider an illustrative monthly budget with invented inputs:

The calculation is $600 ÷ ($60 − $36) = 25 additional orders. Those orders must be additional to what would have happened without the programme. Round a fractional requirement up when planning in whole orders.

Under the same invented monthly inputs, the scenarios are:

Suppose the dashboard credits email with 100 orders. If all extra orders are within that credited group, at least 25 ÷ 100 = 25% must be additional to cover this illustrative budget, assuming the same contribution per order. This is a conditional threshold, not a measured additional share. Check total sales in the comparison below, including purchases the dashboard did not credit.

Margins also change the decision. If contribution falls to an illustrative $12 per extra order, the same $600 budget needs 50 additional orders. A discount can also reduce margin on purchases that would have happened anyway. Add that loss to your assessment; the simple extra-order calculation does not capture it by itself.

How can you check whether email earns that contribution?

For a defined promotional campaign or flow, a randomised no-send comparison can help estimate the additional contribution. Compare total order contribution per person assigned to receive the treatment with contribution per person assigned to the comparison group, over the same buying-cycle window. This is a suggested application of the randomised design used in the ticket-platform study of emailed offers published online in 2016.

Define the audience, treatment, cost and outcome window before sending. Withhold only the promotional treatment being tested; keep unrelated contact consistent and leave service and transactional messages outside the experiment. Include purchases across the business’s measurable sales channels, rather than only orders credited to an email click or open.

Compare the estimated extra contribution for the audience receiving the treatment with the relevant programme cost. Using contribution rather than order counts also accounts for differences in discounts and order margins.

A test of one campaign answers a question about that campaign. It does not establish the value of the entire email programme unless the design actually compares that programme with its absence.

Size the test around baseline purchasing, the smallest effect worth paying for and the precision you need. If orders are sparse or the uncertainty spans your break-even point, the result is inconclusive. A sales increase after a send is not enough by itself: a simultaneous promotion or seasonal change could also affect sales.

Should you start, repair or expand your programme?

Start with a costed programme for a contactable audience. Repair unresolved permission, delivery or measurement gaps before expanding. Increase the budget when credible additional contribution supports the extra cost, considering alternative uses of that money. Treat unclear results as unclear.

The table below is a suggested decision aid, not a ranking of tactics by proven ROI.

A welcome flow for new subscribers or a replenishment email for an appropriate repeat purchase are possible starting points when they match an actual customer need. Neither is automatically the right investment for every store.

If permission is the uncertainty, read whether you can email people who never opted in. If messages are struggling to reach recipients, use the deliverability diagnosis guide before increasing volume.

There is no list-size shortcut in this calculation. A small audience can support a modest programme if the additional contribution covers its cost; a large list does not establish profitability. Set the budget around a customer need and a break-even requirement you can test, then expand on evidence from that programme.