Why email suits online sales
Every other sales channel pays for each person it reaches. Email pays for software and for the time it takes to write, design and test, and then reaches subscribers at almost no extra cost per send. More importantly, it can react to behaviour: a shopper who leaves a basket, browses a category twice or signs up from a pop-up can receive a message within hours that refers to exactly what they did. Those triggered messages, the flows, reach people at the moment they are closest to buying, which is why they tend to produce more revenue per message than broadcast campaigns.
The limits are reach and honesty. Email can only sell to people already on the list, so it cannot find new customers on its own; it converts the interest that paid search, paid social and organic search created. And because its audience is so close to buying, its own report overstates what it adds. A subscriber who was going to reorder this week will click the newsletter and be counted as an email sale. That does not make the channel less valuable, but it means its share of revenue is the wrong number for budget decisions. Consent rules shape the channel too: in the UK, the Information Commissioner's Office guidance on direct marketing sets out when a shop may email people for marketing, and mailbox providers' sender guidelines decide whether the message arrives at all.
Email tactics for online sales
| Tactic | What it does for online sales | Playbook |
|---|---|---|
| Run cart and browse abandonment flows | Reminds shoppers of what they left within hours, which recovers orders other channels already paid for | Abandoned cart flows |
| Earn the first order with a welcome series | Converts new subscribers while interest is highest, with proof and a reason to buy now | Welcome series |
| Plan promotions without training discounts | Mixes product news and content with offers so subscribers do not wait for the next code | Promotional vs transactional email |
| Segment by recency and category | Sends offers matched to what and when each subscriber last bought, cutting irrelevant sends | Segmentation |
| Hold out a share of subscribers | Measures what email adds beyond orders that would have happened anyway | Email metrics and testing |
Recover the orders paid media already bought
A shopper who added a jacket to the basket and left cost something to bring to the site: a click, an impression, a creator fee. The cart flow gives that spend a second chance. Send the first reminder within hours, show the item and the basket, answer the usual hesitations about delivery, returns and sizing, and stop the sequence the moment the order is placed. Browse flows do the same for people who looked at a product or category without adding anything, with a lighter touch. The abandoned cart flows guide covers timing, content and the case for and against incentives. For the sales objective, these flows are the first thing to build because they convert intent that already exists. The cart abandonment rateShare of shopping carts created that did not become an order. shows how much of that intent is left behind.
Turn a subscriber into a first-time buyer
A new subscriber has just shown interest and has not yet bought. The welcome seriesThe automated sequence of emails sent to a new subscriber in the days after sign-up. should use that window: introduce what the shop sells and why it is different, show proof from other customers, answer the questions that stop first orders, and give a reason to act now, which need not be a discount. Read the series on the share of subscribers who place a first order within its window, not on opens. The welcome series guide covers sequence length and content.
Keep promotions from becoming the only reason to buy
A calendar built entirely of offers trains subscribers to wait for the next one, and the shop ends up selling at a discount to people who would have paid full price. Mix campaigns: new products, stories about how things are made, guides that help people use what they bought, and offers at planned moments rather than whenever sales are slow. The promotional versus transactional email guide explains the distinction and the consent rules that apply to each; a receipt or shipping notice is transactional emailAn email triggered by a person's own action or account, such as a receipt, password reset or shipping notice, rather than a marketing send., and loading it with offers can turn it into marketing. Track the discounted share of email revenue; when it climbs, the calendar is doing the selling with margin rather than with reasons.
Send fewer, better-matched campaigns
A subscriber who bought a winter coat last week does not need the coat promotion. Segment by how recently each person bought and what they bought, and send each group the offer or news that fits. Recent buyers might get care guides and accessories; lapsed buyers a reason to return; people who browse but never buy a reassurance message. Fewer irrelevant sends mean fewer unsubscribesShare of delivered emails that led the recipient to opt out., which protects the list the whole channel depends on. The segmentation guide covers how to build the segments.
Measure what email adds
Hold out a random share of subscribers from campaigns, and a share of entrants from each major flow, and compare their orders over the same period with those who received the messages. The difference is the channel's IncrementalityConversions caused by marketing.. The email metrics and testing guide explains how to size and read a holdout. Without one, every decision about frequency and discounting rests on a report that counts orders email did not cause.
Budget posture and KPI targets
Email has almost no media cost, so its budget is the platform, design, copywriting and analysis time. Fund those properly: a shop that spends heavily on paid acquisition and runs email with whatever time is left over is wasting the visitors it bought. The real cost of email is send pressure. Each campaign spends a little of the list's goodwill, and the bill arrives as unsubscribes, spam complaintsShare of delivered emails that recipients marked as spam. and, eventually, poorer inbox placement for every message. Cap campaign frequency at the level where unsubscribe and complaint rates stay flat, and raise it only in steps that are read before the next step.
Growth beyond that cap should come from flows, which send only when a subscriber does something and so add revenue without adding pressure on everyone. Keep a small allowance of time for A/B TestControlled comparison of two variants. work on subject lines, offers and flow timing. Move effort out of a campaign type whose incremental revenue in the holdout is close to zero, and out of discount-led campaigns whose orders come mostly from people who also buy at full price. The list itself is the ceiling: email cannot sell more than its subscribers will buy, which is why the paid channels that grow the list matter to it. How much those channels can pay for each new address is answered in the guide to what an email subscriber is worth, which values a subscriber on the margin the programme adds rather than on attributed revenue.
The tactics in a worked plan
Fallowby Paper, an invented online stationery shop
Fallowby has 48,000 email subscribers and sends three promotional campaigns a week, two of them with a discount code. Email reports £31,000 of a monthly £90,000 in online revenue. There is no browse flow, the cart flow sends one email with 15% off, and 41% of email revenue uses a code.
| Tactic | Before | After |
|---|---|---|
| Cart and browse flows | One cart email with 15% off | Three cart emails, no code until the third, first-time buyers only; a two-email browse flow |
| Welcome series | One discount email | Four emails: story, bestsellers, reviews, then a first-order offer |
| Promotions | Three a week, two discounted | Two a week; one discounted campaign a fortnight |
| Segments | Whole list | Bought in last 60 days, lapsed 61 to 365 days, never bought |
| Holdout | None | 5% of subscribers held out of campaigns; 10% of flow entrants held out |
After three months the holdout shows campaigns add about £6,800 a month in orders that would not otherwise have happened, and flows about £8,900. Reported email revenue falls slightly to £29,500, but the discounted share drops to 22%, revenue per recipient on flows rises from £0.41 to £0.63, and monthly unsubscribes fall from 1,100 to 640.
Fewer sends and fewer codes produced almost the same reported revenue, more of it at full price, and a list that shrinks more slowly. The holdout gave the shop, for the first time, a figure it could compare with paid channels.
KPI set for email online sales
| KPI | What it tells you | Read how often |
|---|---|---|
| Revenue per recipient | What each message earns, comparable across campaigns and flows of different sizes | Per send and monthly |
| Flow revenue as a share of email revenue | Whether growth comes from behaviour-led flows or from more broadcast pressure | Monthly |
| Placed-order rate | Share of recipients who ordered, less distorted by basket size than revenue | Per send |
| Incremental revenue against a holdout | What email adds beyond orders that would have arrived anyway | Quarterly |
| Discounted share of email revenue | How much of email's selling depends on giving away margin | Monthly |
| Unsubscribe and complaint rate | The cost of send pressure on the list and on deliverability | Per send |
Read revenue per recipient beside the unsubscribe rate: a campaign that earns well and loses many subscribers may cost more in future sales than it made. Read the discounted share beside average order value (AOVAverage revenue per order.) and conversion rate (CVRShare of visits or clicks that convert.), because a code can lift both on the day while lowering what each order is worth. The ecommerce email and SMS flows guide maps which flows feed each figure, and the lifecycle email programme playbook sets out the operating routine for reviewing them.
When email is the wrong route for online sales
Email cannot sell to people who are not on the list, so a new shop with few subscribers will get little from it until other channels bring visitors and sign-ups. It is also the wrong place to chase a sales target that requires new customers, because almost everyone it reaches already knows the brand; leaning on it to hit a monthly number usually means more discounts to existing buyers, which raises reported sales and lowers margin. And where subscribers were added without clear consent, sending more will damage deliverability and breach direct marketing rules before it produces orders. In those cases, fix list growthNet change in an email list over a period as a share of its size at the start. and consent first, and look to paid search and paid social for the orders.
