SMS marketing in the EU: how to build a list that actually converts

Why SMS is worth the effort

A text message lands on a device people check constantly, in a channel with almost no competition. No algorithm decides whether it gets seen. No inbox filter files it away. Where email competes with everything else in a crowded inbox, a text arrives on its own.

That is exactly why it is regulated, and the regulation is the reason it still works. Consent requirements keep the channel clean, which is why open and response rates stay high. A business that respects the rules has a channel its competitors have largely abandoned because they could not be bothered to earn the list.

So the question is not whether SMS works. It is how you build a list at the scale that makes it worth running. That is a funnel problem, and it is solvable.

The consent engine: how the list actually gets built

Every SMS programme we run is fed by an acquisition system. Consent is not something you go looking for at the end — it is the output of a funnel designed to produce it. The channels below are how that works in practice, and most programmes combine several.

1. A dedicated landing page with a single job

One page, one offer, one action. The offer is specific and worth a phone number: a discount on a first order, early access, a booking slot, a price list, a callback. The page states plainly that signing up means text messages, from a named business, and how often. Everything else comes off the page.

The conversion rate here is the whole game. A focused page with a real incentive and unambiguous consent wording will out-perform a general "newsletter" box on a homepage by a wide margin, because the visitor understands exactly what they are trading their number for.

2. Paid traffic pointed at that page

Meta and Google ads give you volume on demand. The ad carries the hook; the landing page captures the number. This is the fastest way to reach a list size that makes SMS worthwhile, and it makes the economics measurable: cost per consented number is a number you can watch and improve, exactly like cost per lead.

Targeting matters more here than in most campaigns. You want the audience with a reason to want the offer, because a consented number that never engages drags down deliverability for everyone else on the list.

3. Capture at the point of sale

Checkout, booking confirmation and account creation are the highest-intent moments you will ever get. The person is already buying. A clearly separated, unticked box offering order updates or an incentive in exchange for marketing texts converts well because the value is obvious and the context is honest.

4. Lead magnets and gated content

A genuinely useful download, checklist, calculator or guide, gated behind a form that captures the number alongside the email. This works best when the magnet speaks directly to the problem your service solves, so the list stays relevant rather than becoming a pile of freebie-hunters.

5. Events, in-person and QR codes

A stand, a printed sign, a table card with a QR code that opens a short form. The consent is captured on a real form with the real wording visible, in a moment when the person is already interested. It is slower than a paid campaign and the list quality is usually higher.

6. Text-to-join keywords

A shortcode or keyword people text to opt in themselves. The strongest consent signal available, because the person initiates it from their own device. Best used as a call to action in broadcast, packaging, receipts and social posts.

Run these together and the numbers add up: a landing page converting at a workable rate, paid traffic feeding it, checkout and events adding the warm traffic, and every sign-up recorded with its wording and timestamp. That is a compliant list built deliberately rather than a list you hoped to find.

Campaign structure that gets replies

• Identify yourself in every message. People who do not recognise the sender report the message as spam, and reports damage the channel for everyone using it.
• Include an opt-out instruction in every message. A single word like STOP, plus what it does.
• Set a frequency cap and keep it. The most common cause of complaints is volume, not content.
• Respect quiet hours. A 06:40 text reads as an intrusion regardless of what it says. Send during the working day in the recipient's own time zone, and be stricter at weekends.
• Front-load the value. The first line is all most people see in a notification. If the point is in sentence three, sentence one has failed.
• One call to action per message. SMS is a nudge, not a brochure.

Two-way replies are the point

SMS outperforms email in one specific way: it invites a reply. That is the feature to build around, and most businesses waste it by sending one-way broadcasts from a number nobody monitors.

Decide in advance how keyword replies work. A person replying STOP, END or UNSUBSCRIBE must be removed promptly, automatically and everywhere — not argued with, and not re-added by the next import. A person replying with a question should reach a human within a sensible window, because the alternative is a customer who concluded you do not read your own messages.

When SMS is the wrong channel

• Complex offers. If the decision needs a comparison table, the right job for SMS is to get someone to the thing that explains it, not to explain it.
• Long consideration cycles with several decision-makers. A purchase that needs four stakeholders and a procurement process does not get moved by a text.
• If you cannot staff replies. A two-way channel you do not answer is worse than a one-way channel you do.
• If your list is small and stale. Texting four hundred people who signed up three years ago will produce complaints, not revenue.
• If you cannot prove consent. Then the channel is closed to you for that list. That is the point of this guide.

The test we apply: if this message arrived on the wrong day, to someone in a meeting, would it feel like a helpful nudge or an intrusion? If the honest answer is intrusion, send an email instead.

Where most people get this wrong

The defining mistake is treating consent as a formality to get past rather than the thing that makes the channel work. Bought lists feel efficient because the cost is visible and the penalty is not. But a purchased list fails on its own terms: the people on it did not ask for you, so they do not convert, they complain, and the complaints get your sender profile blocked. You have paid money to make the channel unusable, and handed a regulator the evidence.

The second mistake is treating a STOP reply as an objection to overcome. There is no recovery here. A person who asks you to stop is telling you the relationship is over, and texting them again converts a marketing problem into a legal one.

The third is measuring the channel by volume sent. "12,000 messages delivered" says nothing about whether the channel works. What matters is how many real conversations were started and how many became customers.

The short version

• SMS is the highest-response channel most businesses can access, and it is worth the effort of earning the list.
• The list is a funnel output. A dedicated landing page, paid traffic, checkout capture and lead magnets are what produce it.
• Consent must name SMS, be separate from other terms, and be as easy to withdraw as to give.
• Log the number, wording, action, source and date. Consent you cannot demonstrate is consent you do not have.
• In B2B, the position varies by market. The UK allows marketing to corporate subscribers without prior consent under PECR; some EU states carve out a similar exception and others do not. We confirm it per country before sending.
• Buying, scraping or inheriting numbers never works — it destroys the channel rather than building it.