⚡ Quick answer: Email still returns more than any other channel — $36-42 for every dollar spent, rising to $45 in retail. But the rules changed in 2024: Gmail and Yahoo now require SPF, DKIM and DMARC from senders, and a spam complaint rate above 0.3% quietly kills your inbox placement. The part most owners miss: a purchased list, or one built without consent, isn’t merely ineffective any more — it actively damages the domain you send from.
The most common way this reaches us is “our newsletter stopped working.” Usually the problem isn’t the copy or the button — the emails simply aren’t arriving, and the report says nothing about it, because they’re all counted as sent.
The reason is unglamorous: the list belongs to you. Instagram’s algorithm can halve your reach overnight, an ad account can get suspended without explanation, but a list of addresses sits in your account and isn’t going anywhere.
Returns vary a lot by sector, and that’s worth building into expectations before you start:
| Sector | Return per $1 |
|---|---|
| Travel and hospitality | $53 |
| Retail and ecommerce | $45 |
| Marketing and advertising | $42 |
| Software and technology | $36 |
| Media and publishing | $32 |
These are market averages, and they’re a reference point rather than a promise. A business emailing a three-year-old list once a quarter won’t see $36, or $10.
The single most telling figure in this whole topic: automated emails generate 37% of all email-driven sales while accounting for just 2% of sends. A handful of flows configured once outperform every manual campaign combined.
The logic mirrors retargeting: the email arrives right after the person did something, not when it’s convenient for a marketer to hit send. The difference in relevance is enormous.
The minimum set of flows worth starting with is three or four, no more:
This is the part most small businesses skip, then wonder why nothing arrives. Since February 2024, Google and Yahoo enforce technical requirements on senders, and the formal “bulk sender” threshold is 5,000 emails a day to personal accounts. SPF and DKIM, though, are recommended for everyone regardless of volume.
| Requirement | What it means in practice |
|---|---|
| SPF | A DNS record listing the servers allowed to send mail from your domain |
| DKIM | A cryptographic signature proving the message wasn’t altered in transit |
| DMARC | The policy tying the first two together; p=none minimum for bulk senders |
| One-click unsubscribe | Mandatory for promotional mail, must be honored within 48 hours |
| Spam complaints | Under 0.3% — and that’s the enforcement line, not a safe target |
The complaint threshold deserves its own moment. Google says outright that 0.3% is where enforcement begins, not a goal. The working ceiling for stable delivery is under 0.1%. Between those two numbers, deliverability degrades silently: emails don’t bounce with an error, they just start landing in Promotions and Spam more often, and your report won’t show it.
Three complaints per thousand emails and you’re over the line.
Repairing domain reputation takes two to six weeks of disciplined sending: cut volume down to your most engaged subscribers, fix whatever caused the complaints, then rebuild gradually. It’s slow and unpleasant, which is why it’s cheaper never to get there.
Buying a list is the fastest way to lose a domain, and that’s worth stating bluntly, because “let’s buy 20,000 addresses and start” sounds temptingly cheap. People who don’t know you file spam complaints at rates far above the 0.3% threshold, and it’s your domain that pays, not the list vendor. At Netloria we don’t take this kind of work at all — not on moral grounds, but because the outcome is predictably bad.
What works instead:
List growth rate is a poor metric on its own. A thousand subscribers who genuinely open your emails are worth more than ten thousand dead addresses dragging your domain reputation down.
A few years ago opens were the primary measure of a campaign. Apple’s Mail Privacy Protection broke that: Apple Mail preloads images in messages, so the system records an open even when nobody looked at the email. In lists with a heavy iPhone share, these phantom opens can reach half the total.
What to watch instead:
Market averages for reference: opens around 19-21%, clicks 2.1-2.6% depending on source and sector. If your clicks sit at 2%, you’re roughly normal, and chasing somebody else’s 5% is pointless without knowing what kind of list produced it.
There’s no universal answer, but there is a working principle: frequency should match how often you actually have something worth saying. A clothing store with new collections can email weekly. A law firm at that pace becomes an irritant within a month.
For most small service businesses the workable rhythm is two to four emails a month. Less than once a month is also a problem: people forget you, and the next email reads like spam from a stranger, even though they subscribed themselves six months ago.
The best indicator isn’t instinct, it’s the list’s reaction. Unsubscribes and complaints rising after you increased frequency means you overdid it. Clicks holding steady with flat unsubscribes means you can try more often.
Segmentation sounds complicated, but two or three groups are plenty to start. Someone who already bought and someone who just subscribed are different people with different questions. The first doesn’t need convincing that you’re trustworthy; the second shouldn’t be offered an add-on for something they don’t own yet.
The simplest split we use at Netloria that pays off almost immediately: active (opened in the last 3 months), dormant (nothing in 3-6 months), and customers. Three segments, three different tones — and clicks usually rise without changing a word of the copy, simply because the message finally reaches the right audience.
The dormant segment needs regular pruning. An address that hasn’t opened a single email in six months won’t produce a sale, but it does drag down the engagement statistics Gmail looks at when deciding whether to put you in the inbox or in Promotions.
Service companies find this question hardest. A store has new arrivals and sales, but what does a law firm, a dental practice or a construction company send? So the newsletter either never launches or turns into “season’s greetings” four times a year.
The workable approach is to write about what you already explain to clients in meetings. Every business has a set of questions that repeat dozens of times a year: how long does this take, why is it cheaper down the road, what goes wrong if you cut corners. Those are ready-made email topics that genuinely get opened, because they answer a question already in the reader’s head.
The format can be as plain as it gets: a short 200-300 word email with one idea and one link. Long designed templates with banners and three columns aren’t a requirement — often the opposite, a simple text email from a named person at the company gets opened more readily than an advertising layout.
A telling case from our own work: an online store came to us with a list of 14,000 addresses built over three years with no confirmation step, complaining that the newsletter “stopped working.” The check turned up a familiar picture — SPF partially configured, DKIM missing entirely, spam complaints around 0.4%, above the enforcement line. Emails were genuinely being sent and even marked as delivered in the report, but Gmail was quietly filing most of them into spam, and the owner couldn’t see it because he was watching only sends and opens, and the opens were coming from the portion of the list on other mail providers. The fix was unpleasant: our team at Netloria configured authentication, then cut roughly 9,000 addresses that hadn’t opened anything in a year. The list dropped to 5,000 — and revenue from the channel started growing only after that, because the remaining emails finally reached people.
The takeaway is obvious but hard for an owner to accept: a smaller live list earns more than a large dead one.
Order matters here, and the technical part comes first — before copy, before email design, before choosing a platform.
Start by checking whether SPF and DKIM are configured for your domain. It’s a one-time job, under an hour for a specialist, and nothing else matters without it. Next, a separate domain or subdomain for sending if you’re planning real volume: that way your company’s main email isn’t collateral damage if something goes wrong with the newsletter. Then a signup form with an honest promise and double opt-in. And only after all that — a welcome sequence and a first send to a small segment, to watch how people react.
Your list will be small for the first two or three months, and that’s fine. Email is a channel that pays off over distance, and its real advantage is compounding: today’s 200 subscribers become 2,000 in a year, and they’re yours rather than rented from an algorithm.
If you want subscription forms and email flows built into the site properly from the start rather than bolted on afterwards — take a look at how we handle promotion. We’re a web studio based in Ukraine, and domain configuration plus form integration comes as part of the work rather than a separate invoice.
