Email marketing and automation are two different jobs that most platforms bill you for as one subscription: email marketing is you writing and sending a campaign, while automation is the system firing pre-built sequences — welcome series, cart recovery, re-engagement — whenever a contact's behaviour triggers them. The distinction matters because automation keeps earning after you've stopped working, and it's also where usage-based pricing on platforms like Mailchimp climbs fastest as your list grows.
TL;DR:
- Email marketing is you writing and sending a campaign; automation is pre-built logic that sends without you touching a button — most platforms charge for both the same way, by contact count.
- The seven email types every programme needs: welcome, newsletter, promotional, transactional, cart/activity recovery, re-engagement, and milestone emails.
- The DMA UK's Marketer Email Tracker has long put email ROI at roughly £42 for every £1 spent — the channel works, the question is what you're paying to run it.
- Sending cost and platform cost are separate line items: Amazon SES charges $0.10 per 1,000 emails ($0.16 for new accounts), while the platform fee is the part that actually scales with your list.
- AI in email tools mostly handles subject-line testing and send-time optimisation, not strategy — list hygiene and segmentation logic still have to be built by a person.
Table of Contents
- Email Marketing and Automation Aren't the Same Spend, and Confusing Them Costs You Money
- The Seven Email Types Every Automation Programme Actually Needs
- What Marketing Automation Looks Like in Practice: Five Examples
- Does Email Marketing Actually Work, or Is That a Vendor Talking Point?
- What AI-Powered Email Marketing Actually Does (and Where It's Oversold)
- How Much Automated Emails Really Cost: Platform Fees vs Sending Fees
- Where the Per-Contact Pricing Model Actually Breaks
- Get Email Marketing and Automation Without the Per-Contact Bill
- Frequently Asked Questions
- Sources
Email Marketing and Automation Aren't the Same Spend, and Confusing Them Costs You Money
Marketing, in the broadest sense, is anything that moves a stranger closer to buying from you. Email marketing is one channel for doing that — campaigns you write and schedule yourself. Automation is a delivery mechanism inside that channel: a sequence built once, triggered by something a contact does (signs up, buys, goes quiet), that then runs without you. Most platforms sell both under a single monthly price tied to your contact count, which is exactly why the two get muddled. You can run a perfectly good email marketing programme with zero automation, sending one newsletter a month by hand. You can also run automation with almost no "marketing" in the traditional sense — a password reset, an order receipt, a shipping update. They're billed together but they do different work, and if you're evaluating email automation software, the sequences matter more to your growth than the campaigns do, because they run whether or not you remember to log in.
The Seven Email Types Every Automation Programme Actually Needs
Every mature email programme, regardless of platform, ends up sending some version of the same seven email types. Missing one usually shows up later as a revenue leak you can't quite locate.
| Email type | What triggers it | Job it does |
|---|---|---|
| Welcome | A new signup | Sets expectations, often the highest open rate you'll ever get from that contact |
| Newsletter | A schedule | Keeps you top of mind between purchases |
| Promotional | A manual decision | Drives action on a specific offer or deadline |
| Transactional | A purchase or account action | Confirms an order, a password reset, a shipment |
| Cart or activity recovery | Inactivity after a specific action | Recovers revenue that's already half-committed |
| Re-engagement | Long-term inactivity | Reactivates or cleans the contact off your list |
| Milestone | A date (birthday, renewal, anniversary) | Gives a reason to email that isn't a sale pitch |
Transactional and cart-recovery emails are the ones businesses underbuild — they're treated as an afterthought when they're usually the highest-converting emails you send, because they're relevant to something the person just did.
What Marketing Automation Looks Like in Practice: Five Examples
- Cart or form abandonment. Someone starts checkout or fills a form and leaves; an email fires after a set delay, often with the exact item or offer they were looking at.
- Lead nurture. A new subscriber gets three to seven emails over two to three weeks, each building a case, before a sales email ever appears.
- Post-purchase upsell. A customer buys, and 10–30 days later receives a relevant add-on offer timed to when they'd realistically need it.
- Win-back. No opens or clicks for 90–180 days triggers a "we miss you" send, followed by a suppression if there's still no response — this is also how you keep your list, and your sender reputation, clean.
- Onboarding drip. A new account gets a scheduled run of setup emails rather than one long welcome message nobody reads in full.
If you want a deeper look at how long these sequences should run and when to stop sending to someone who isn't responding, the mechanics are covered in drip campaigns: build sequences that stop when they've worked.
Pro Tip: Before building any automation, pull the actual open-to-click ratio from your last 90 days of manual sends and use that as your baseline — not an industry average. If your real click rate on a normal campaign is 1.5%, an automation promising a 12% click rate from a template is telling you more about the benchmark than about your list.
Does Email Marketing Actually Work, or Is That a Vendor Talking Point?
Yes, with a caveat that most vendor pages leave out. The DMA UK's Marketer Email Tracker has repeatedly put email marketing's return on investment at around £42 for every £1 spent, which is why it remains one of the highest-ROI channels available to small businesses. But that figure assumes a list that's actually opted in and reasonably engaged — a purchased or scraped list produces complaints, not revenue, and under UK and EU rules enforced through the ICO's PECR guidance you need clear consent before sending marketing email in the first place.
£42 for every £1 spent — that's the return the DMA UK has tracked for email marketing, and it's still the reason the channel outlasts every "email is dead" headline.
There's a second complication worth knowing: open rates have been unreliable as a metric since Apple introduced Mail Privacy Protection in 2021, which pre-loads images for a large share of Apple Mail users regardless of whether they actually read the email. If you're judging a campaign's success by open rate alone, you're partly measuring Apple's infrastructure, not your subscribers' attention. Clicks and conversions are the figures that still mean what they say.
What AI-Powered Email Marketing Actually Does (and Where It's Oversold)
"AI-powered" on an email tool's feature list usually means one of a handful of specific things, not a general intelligence running your strategy.
- Subject-line generation and testing. The tool proposes variants and picks a winner based on early open or click data — useful, but it's testing wording, not strategy.
- Send-time optimisation. The system times delivery to when a specific contact has historically opened email, rather than sending your whole list at 9am.
- Predictive segmentation. Contacts get scored or grouped by likely behaviour (churn risk, purchase likelihood) based on past activity, so you can target the automation differently.
- Content personalisation. Product recommendations or dynamic blocks are swapped per contact using past purchase or browse data.
None of it replaces the decisions a person still has to make: which segments exist, what the automation's logic branches on, and when to stop emailing someone who's gone cold. Treat AI features as a tuning layer on a structure you've already built, not the structure itself.
How Much Automated Emails Really Cost: Platform Fees vs Sending Fees
There are two separate costs here and almost every pricing page blends them into one number, which makes comparison harder than it should be.
| Cost component | What it covers | Typical range |
|---|---|---|
| Platform fee | Campaign builder, automation engine, forms, landing pages | Usually scales with contact count on subscription tools |
| Sending fee | The actual delivery of each email | Amazon SES charges $0.10 per 1,000 emails, or $0.16 per 1,000 for new accounts |
| Hosting (self-hosted tools) | Server to run the software | Roughly €5–15 a month for a small VPS |
On a subscription platform, the platform fee and the sending cost are usually the same bill, and that bill climbs in tiers as your contact count rises — whether or not those contacts are actively buying from you. On a self-hosted tool, the platform is a one-off licence and you route sending through your own provider, so the sending fee and the licence fee are permanently separate. ReplaMail is built on that second model: a one-time $299 licence for a single company (or $899 for agencies managing multiple brands), with 12 months of updates included and an optional $49 a year after that if you want to keep updating. You still need somewhere to send from — Amazon SES, SMTP2GO, Resend, SendGrid, Mailgun, Postmark, or your own SMTP — and a small server to run it on.
Where the Per-Contact Pricing Model Actually Breaks
Per-contact pricing works fine at small list sizes and becomes expensive precisely when your business is succeeding — more subscribers, more revenue, and a bill that grows alongside both. Most mainstream platforms, including Mailchimp, price in tiers by contact count, so crossing a tier threshold (even by a handful of contacts) can jump your monthly cost rather than scaling it smoothly. Agencies feel this hardest: running five client brands on a per-contact platform usually means five separate subscriptions, because most tools aren't built to separate brands cleanly within one account. If you're actively comparing structures, marketing automation software: the real buying decision and email marketing tool: what actually fits your list and budget both walk through where the break-even sits for different list sizes.
A list that costs $20 a month at 2,000 contacts can cost well over $100 a month at 20,000 — the automation logic you built didn't change; the contact count did.
Get Email Marketing and Automation Without the Per-Contact Bill
If the problem you're solving is a platform bill that keeps rising while your automations stay exactly the same, ReplaMail is built for that specific situation: a self-hosted email marketing system you buy once for $299 (or $899 for agencies running multiple brands), with no per-contact fee ever, because you route sending through your own Amazon SES, SMTP2GO, SendGrid, Mailgun, Postmark or SMTP account. It includes a drag-and-drop campaign builder, branching automations, signup forms, landing pages, double opt-in, and bounce and complaint handling — covering everything described above.
It isn't a sending service, so you'll need an SMTP provider and a small server, roughly €5–15 a month. It also doesn't do SMS, social posting, A/B testing, Shopify or WooCommerce integration, backups, RSS-to-email or CRM — if any of those are central to how you work, check the spec sheet (ours included) before you commit, because the one-time price only makes sense if the feature list actually covers what you send.
Frequently Asked Questions
How much do automated emails cost?
There are two separate costs: a platform fee for the tool that builds and triggers the automation, and a sending fee for actually delivering each email. Subscription platforms usually bundle both into one tiered, per-contact bill, while self-hosted tools separate them — a one-time licence plus a sending provider like Amazon SES, which charges $0.10 per 1,000 emails ($0.16 for new accounts).
What are the 7 types of emails?
Welcome, newsletter, promotional, transactional, cart or activity recovery, re-engagement, and milestone emails. Together they cover everything from a first signup to a long-dormant contact, and most automation programmes end up building some version of all seven over time.
What is AI-powered email marketing?
It usually refers to a specific feature, not a general strategist: subject-line testing, send-time optimisation based on a contact's past open history, predictive segmentation, or dynamic content personalisation. It tunes an automation you've already built rather than designing the automation's logic for you.
Does email marketing really work?
Yes — the DMA UK's Marketer Email Tracker has repeatedly put email ROI at around £42 for every £1 spent, making it one of the highest-return channels available to small businesses. That figure assumes a genuinely opted-in, engaged list; a purchased or scraped list produces complaints and bounces rather than revenue.
What's the real difference between email marketing and marketing automation?
Email marketing is a campaign you write and send yourself; marketing automation is a sequence built once that fires on its own when a contact's behaviour triggers it, such as abandoning a cart or going inactive. Most platforms price both the same way, by contact count, which is why they get treated as one thing when they're really two different jobs.
Sources
- DMA Marketer Email Tracker | Data & Marketing Association (DMA UK)
- Guide to PECR: electronic mail marketing | Information Commissioner's Office (ICO)
- Amazon SES Pricing | Amazon Web Services
- Mailchimp Pricing | Mailchimp