Is SMS marketing worth it for ecommerce?

A text is read within minutes. That is the whole value, and it means the messages worth sending are the ones where minutes matter.

Limited releases and restocks are the clearest case. When a product sells out quickly, the customers who hear first are the ones who buy, and an email sitting unopened for four hours is not competitive with a message read immediately.

Delivery and order updates are the second. These are the messages customers most want, they generate goodwill rather than annoyance, and they reduce support enquiries about where an order is.

Abandoned checkout recovery is the third, and it works because the window is short. A single message shortly after someone leaves a checkout recovers orders at a rate that justifies its cost, provided you are not also sending three emails about the same basket.

What does not work is a weekly promotional text. It costs money on every send, it competes with messages from people the customer actually knows, and it produces opt-outs that remove the person from the channel permanently.

The general shape is consistent across markets: you need clear, specific, provable consent to send marketing texts, it cannot be bundled into a general signup without the person understanding what they agreed to, and the record of that consent needs to be kept.

Beyond that the detail differs. Countries and regions vary on what counts as adequate consent, whether an existing business relationship changes anything, what identification a message must carry, and what hours messages may be sent in. Sending across borders means meeting the rules of the recipient's location rather than yours.

Opt-out has to work immediately and through the standard keywords people expect, and it has to be honoured across every list you hold rather than only the one the message came from.

Because the rules are enforceable and the numbers of messages are large, this is one area where getting a proper answer for your markets before launching is genuinely cheaper than fixing it later. Treat the platform's default settings as a starting point rather than as compliance.

Email is close to free at the margin, so a campaign to the whole list has almost no downside beyond attention. Text messages have a per-message cost, sometimes with different rates by country and by message type, so sending to people who will not buy has a direct price.

That pushes the channel towards segmentation. Sending to the customers most likely to act, rather than to everyone, is not a nicety here; it is what makes the arithmetic work.

Calculate it per campaign. Take the number of recipients, the cost per message, the orders it produced and the contribution on those orders. A message with a modest response rate to a small, well-chosen segment usually beats a broad send with a larger absolute number of orders.

The same applies to length. Messages over a certain length are billed as more than one, so a habit of writing long texts multiplies the cost of the whole programme without improving it.

Set a frequency cap and hold to it. A small number of promotional messages a month, plus transactional ones, is a rhythm most customers tolerate. There is no recovering a subscriber who opted out, so protecting the list matters more than any individual campaign.

Coordinate with email rather than duplicating it. A customer receiving the same offer by both channels within an hour learns that one of them is redundant, and they will unsubscribe from whichever annoys them first. Decide which channel owns which kind of message.

Respect quiet hours in the recipient's own time zone, not yours. A message at seven in the morning in another region is the fastest way to generate complaints regardless of how good the offer was.

And give people a way to reduce frequency rather than only to leave. A preference option that keeps them on order updates while stopping promotions retains a customer who would otherwise have opted out of everything.