SMS marketing for ecommerce: what works once the novelty wears off

It was never the channel being magic. It was being one of very few brands in a quiet inbox, and that part is over.

Ecommerce SMS marketing still works, but only for messages a customer would welcome in the middle of dinner. In other words they need to be properly triggered, time-critical and personal. The brands keeping their results send fewer promotional text messages, let customer behaviour trigger most of the rest, and judge every send by opt-outs as well as revenue.

This guide is for teams that are already sending. It covers what still works in ecommerce SMS marketing, what to stop doing, and how to tell the difference in your own numbers.

Why SMS stopped working the way it used to

SMS had one real asset early on, and it wasn't the open rate. It was scarcity. When most of a customer's text messages came from friends, family and the odd delivery driver, a message from a brand they liked stood out by default. You didn't need a brilliant offer. You just needed to be there.

Then every brand they had ever bought from started sending text messages. The fashion label, the coffee subscription, the pet shop, the furniture retailer they ordered one lamp from. Their usually quiet inbox filled up, and your marketing advantage went with it. Nothing about your SMS messages got worse. You're just trying to talk in a loud room.

It's worth being fair here, because the gloomy version of this story is wrong too. SMS still gets more attention than anything else you own. People read their text messages. What changed is how the channel responds to misuse. Email ignores you when you get it wrong: the message sits unread, the customer scrolls past, and you get another go next time. SMS punishes you. A badly judged text lands on the lock screen next to a message from someone's mum, and the reply is STOP. That subscriber is gone, and they aren’t coming back.

So the question for year two isn't whether SMS marketing works. It does. The question is which messages have earned the right to interrupt someone’s day, and how many of them you can send before that right runs out.

What SMS is good at, and what it isn't

SMS is good at four things, and they tend to work well together: short messages, time-critical messages, personally relevant messages and messages the customer already expects. “Your parcel arrives today between 10 and 12” ticks all four. “Our autumn edit has landed” ticks none of them.

The most useful way to think about the benefits of SMS marketing is as a division of labour with email marketing. Email is like a shop window, with room for images, stories, prices, and a product grid people can browse. SMS is a tap on the shoulder. One fact, one action, no browsing. Brands get into trouble when they ask the tap on the shoulder to do the shop window's job. That holds across retail SMS marketing generally, not only for online-only brands.

SMS is poor at:

• Browsing and discovery. Nobody wants to scroll a catalogue through a link in a text.

• Anything that needs more than a sentence.

• Anything the customer didn't already care about. A new collection, a brand story, a blog post: all fine in email marketing, all intrusions by text.

The practical test is simple, and it settles most arguments in planning meetings. Would this message be welcome if it arrived while they were at dinner? A delivery update, yes. A back-in-stock alert for the trainers they've been waiting for, probably. A weekend discount sent to the whole list, no. If the answer is no, the message belongs in email, where it can wait politely until they're ready.

Promotions can still go by SMS. They just have to clear a higher bar than everything else, which is what the next section is built around.

SMS marketing examples that still work

These six SMS marketing examples keep working after the novelty has gone. The table shows what triggers each one, when it should go out, and why customers still put up with it.

Look at the trigger column. Five of the six are set off by something the customer did: they filled a cart, asked for an alert, placed an order, watched a product or bought something that runs out. The sixth is set off by your marketing calendar. That's the exception, and it should be the only one.

This is the pattern behind most declining programmes. Triggered SMS messages are read as service, because the customer started the conversation. Calendar-driven SMS marketing campaigns are read as advertising, because you did. 

A few practical notes on the triggered five:

• Abandoned cart: one text, no more. If email already handles the cart flow, SMS should be the fast first nudge or the last one, not both.

• Shipping and delivery: these transactional text messages also take pressure off customer support. A customer who knows where their parcel is doesn't need to pester you.

• Replenishment and post-purchase: the timing should come from what was bought. A tub of protein powder and a winter coat have very different next moments, which is why post-purchase automation that is not generic beats one follow-up for every order.

• Price drop: only for items someone clicked watched or saved. A price drop on something they glanced at once doesn't cut it.

So if you are looking for what to send by SMS instead of generic promotions, it is the five triggered messages above, plus the rare sale that genuinely deserves a text.

How often should an ecommerce brand send SMS?

Start with two promotional SMS campaigns a month, on top of the text messages customers trigger themselves. Add a third only if the opt-out rate on each send stays close to your usual level, and if a send draws roughly double your normal opt-outs, treat it as where you went too far.

Frequency is the biggest lever in SMS marketing. The platforms do publish frequency advice, and it tends to land on the send more side, which is what you would expect from businesses that charge by the message. Two SMS marketing campaigns a month will feel cautious if you currently send weekly. It's meant to. Treat it as the ground level which you can earn your way up from, not a ceiling you bargain down from.

Opt-out rate is the number that tells you where the boundary is, because an SMS opt-out costs more than an email unsubscribe. Each phone number on your list was usually collected once, often in exchange for a discount, and when someone texts STOP you can't text them to ask them back. Email has more ways back in: the next order confirmation, an account sign-up, the newsletter box at checkout. A lost SMS subscriber is a cost you paid to acquire and can't easily pay again.

So track opt-outs per send across your SMS campaigns, not per month. A monthly average hides the one campaign that did the damage. Compare each send with your own trailing average rather than any benchmark, because benchmarks for this metric disagree with each other.

Triggered SMS campaigns sit largely outside this budget. Nobody opts out because their parcel is on its way.

Who you send to, and the list that doesn't churn

Consent quality beats list size. A phone number given with a clear promise ("we'll text you when your size is back and when your order ships") behaves completely differently from one harvested behind a discount pop-up. The first customer knows what they signed up for and gets it. The second signed up for a code, used it, and now treats every text as the price they are still paying for it. Guess which one opts out first.

So when you collect phone numbers, say what you'll send and roughly how often, then keep to it. 

Segmentation matters more in SMS than anywhere else, because SMS punishes irrelevance faster than any other channel. In email, a badly targeted message costs you an open. In SMS it costs you the subscriber. Three inputs help you out here:

• Purchase history: what they bought, and what usually follows it.

• Category interest: what they browse, save and watch.

• Recency: how long since they last bought or engaged.

Recency is the one most skip, and it's where predictive scoring to spot who is drifting away can come in very handy. A customer who has gone quiet doesn't need your fortnightly promotion. They need a reason to come back, or a break. None of this works if purchase, browsing and consent data live in separate tools, which is why the data layer comes before the clever tactics.

On compliance, keep it simple and keep it strict. Get clear consent for marketing text messages, make opting out easy in every message and identify yourself honestly as the sender. Rules differ by country, so check the specifics for each market you send to, and treat any phone number without a recorded consent as off limits.

Measuring SMS honestly

Last-touch attribution flatters SMS more than any other channel. The text usually arrives at the end of the journey, often after the customer has already decided, so it gets all the credit for a sale it didn't cause. An abandoned cart text sent to someone who was coming back anyway still shows up as recovered revenue. 

This isn't a criticism of SMS marketing. It's a property of what SMS marketing is all about. A message designed to arrive at the moment of decision will always look decisive to a model that credits whatever happened last.

Three measures give you a better picture though:

• Incremental revenue. Hold back a random slice of a segment, say one customer in ten, from promotional SMS campaigns for a few weeks, then compare revenue per customer between the two groups. The difference is what SMS actually added.

• Opt-out rate next to revenue. Report them in the same table. SMS marketing campaigns that made money and cost you a big slice of your list didn't make money.

• Performance per segment, not per send. A send-level report averages your best customers with your least interested ones. Split it, and you'll usually find one segment carrying the results while another opts out.

If a holdout test isn't possible yet, at least tighten the rules your reports use. Count SMS clicks rather than sends, and keep the attribution window short, so a text from last week can't come along and claim today's order. Working out which channel actually earned the sale comes down to choosing those rules deliberately rather than inheriting the defaults.

What to look for in an SMS setup

This section names no names. The right SMS marketing platform depends on what you already run, and a framework will serve you longer than our opinion of anyone. Four questions will help you get it right.

1. Does SMS share a customer profile with your other channels? If your SMS marketing software keeps its own list, separate from email and your store data, every frequency and segmentation decision above gets harder.

2. Can a flow switch channel partway through? For a mature programme, the most useful SMS marketing platform is one where a journey can check consent, preferences and past messages, then choose email, SMS or something else at each step.

3. Can you see SMS revenue next to everything else? Look for reporting that puts SMS in the same view as email and your other channels, with control over which interactions count, so you compare like with like rather than each tool's flattering version of itself.

4. Where do replies go? Customers reply to texts, often with questions. Check whether replies can reach customer support or trigger a sensible next step, rather than disappearing into the void.

Where Manago AI fits

Manago AI is a customer engagement platform rather than a standalone SMS marketing platform. SMS runs as one channel inside it, alongside email, WhatsApp, push and on-site engagement such as pop-ups and forms. Behaviour, purchases, engagement, preferences, consents and predictive signals sit in a single customer profile, so your SMS messages work from the same picture as every other channel rather than from a separate list.

The part that matters for this article is what happens inside a flow. In Manago AI Workflows, SMS, email, WhatsApp and push can sit in one customer journey, and a step can check conditions before anything is sent: whether the contact has opted in to email, which channels and frequency they chose in the Customer Preference Center, whether a text has already been delivered, or how likely they are to churn. That means you can build the flow to use email when email will do, keep SMS for the moments that need it, and stop a second text reaching someone who has just had one. It's the structural answer to the frequency problem: fewer texts, chosen by rule rather than habit.

Replies become part of the journey too, instead of landing in an inbox nobody in customer support checks. A customer can reply with a keyword, and that reply can send a personalised email, trigger a mobile push notification, apply a segment or stop an irrelevant follow-up. Back-in-stock and price-drop events can start a workflow, which covers two rows of the table above.

On measurement, Revenue Analytics shows SMS alongside email, WhatsApp, push and ad channels, and lets you decide which SMS interactions count, for example crediting clicks while excluding sends.

For the feature-level view, see how Manago AI handles SMS.

The questions that come up once SMS is running

Send less, mean more

Ecommerce SMS marketing didn't stop working. The conditions it worked under changed: the quiet inbox filled up, and the channel started charging for every message that didn't deserve to be there. The brands still doing well with it aren't the ones with the cleverest copy. They're the ones that send less, let customers trigger most of their text messages and read opt-outs as carefully as revenue. If your numbers have slipped since year one, that's where to start. Cut the sends nobody asked for, keep the ones they did, and measure what's left honestly.