Your SMS Programme Is a Discount Megaphone (The 7.6% of Sends That Make 45% of the Revenue)

Klaviyo published SMS benchmarks this year drawn from more than 183,000 accounts, and one line in it should reorganise how most brands run the channel. Flows are 7.6% of all SMS sends. They produce 45.2% of all SMS revenue. Nearly half the money comes from under a tenth of the volume.

Almost every DTC brand I look at has that ratio inverted. The SMS programme is a campaign calendar with a couple of flows switched on as an afterthought, usually abandoned cart and a welcome message, both inherited from the email setup and never rewritten for the medium. All the attention, all the copy time and all the send volume goes to the half of the channel that earns least per message. Figures below are in US dollars because that is how the source data is reported.

The numbers that should change your send calendar

Put campaigns and flows side by side and the gap stops being a nuance and starts being a strategy. On click rate, campaigns average 5.6% against 9.65% for flows. On order rate, the number that actually pays, campaigns average 0.27% and flows average 1.9%. That is seven times the conversion from the same list, the same brand and the same product catalogue. The only difference is whether the subscriber did something that caused the message.

Revenue per recipient makes it starker. In health and beauty, campaigns return $0.32 per recipient and flows return $2.09. In food and drink it is $0.34 against $1.83. In home and garden, $0.32 against $3.74. Whatever the category, the multiple sits somewhere between five and twelve.

That last point is the one worth sitting with. If your SMS reporting shows a healthy revenue number driven mainly by campaigns, a meaningful share of it is cannibalised demand wearing a discount code. You are paying per message to pull forward purchases and compress your own margin, and the dashboard reads it as channel performance.

Why the discount megaphone feels like it works

SMS gives you the fastest feedback loop in the stack. You press send, and within twenty minutes there is revenue on the dashboard. Nothing else in marketing does that. Paid media takes days to read, email takes hours, a lifecycle rebuild takes weeks before it shows anything. SMS pays out while you are still looking at it.

That speed is exactly what makes it dangerous. A channel that rewards you instantly will train you to press send more often, and because each individual send does produce revenue, nothing in the reporting tells you to stop. The cost shows up in two places that are both one step removed from the send report: your list, and your discount floor.

Median 30-day subscriber retention sits at 93.1%, which sounds fine until you annualise it. Roughly 7% of your list leaves every month at the median, and the brands at the 90th percentile hold 99.1%. The difference between those two numbers over a year is most of your list. Every promotional send with nothing behind it is a small withdrawal from an asset you paid to build, and acquisition is slow: the median brand adds new SMS subscribers at 1.06% of new orders.

The four triggers that actually earn

Build these before you add another slot to the promotional calendar. They are ordered by how fast they pay back.

1. Back in stock

Click rates of 36 to 58% and conversion of 7 to 13.8%. Nothing else comes close, and the reason is structural rather than clever: the subscriber asked to be told, the product is already chosen, and the window genuinely matters because stock is finite. This is the one flow where SMS beats email on merit rather than on novelty, because minutes are the whole value. If you only ever build one SMS flow, build this.

2. Replenishment timed to consumption

For any consumable, the reorder message is worth more on SMS than on email because it is a one-tap decision on a product they have already judged. The work is not the copy, it is the timing. Pull actual days-to-reorder from your Shopify data per SKU rather than using the 30-day default, and fire a few days before the jar runs out rather than a few days after.

3. Shipping exceptions, not shipping updates

Routine tracking notifications belong on email or in the carrier flow. The exception does not. A delay, a failed delivery, a missed collection window: these are the moments a customer is about to open a support ticket or start a chargeback, and a message that gets ahead of it costs you one SMS and saves a ticket, a refund request and sometimes the relationship.

4. Cart and browse, rewritten for the medium

Most brands have these switched on and have never rewritten them. The email version opens with a subject line, carries an image and has room to re-sell the product. The SMS version has one line to work with and arrives in the same thread as the recipient's family. Strip it to the product, the reason they stopped and a link. Do not include a discount on the first touch, because a cart flow that always discounts teaches the list to abandon on purpose.

The part nobody prices: consent

SMS is the one marketing channel where a list-building shortcut carries a per-message legal price. TCPA statutory damages run 500 to 1,500 US dollars per message, and because the exposure multiplies by volume rather than by incident, a single badly sourced segment is a class action rather than a fine.

The change most brands have not absorbed is the FCC's one-to-one consent ruling from January 2025, which ended co-registration. If a number came from a lead-generation page that listed several brand partners, that is not consent for your brand, whatever the vendor told you when they sold you the file. Each brand now needs its own clear, conspicuous disclosure between itself and the subscriber.

The mechanics are specific and cheap to get right. Your opt-in form itself has to carry the brand name, what kind of messages you will send, an estimated frequency, the message and data rates disclosure and the STOP instruction. STOP has to be honoured within ten business days under TCPA, and GDPR expects withdrawal to be as easy as giving consent, so process it immediately. Quiet hours mean nothing before 8am or after 9pm in the recipient's local time, which matters more than operators expect once a list spans time zones. For EU and UK subscribers you need explicit consent and a lawful basis, not a soft opt-in inherited from an email signup. This is general information rather than legal advice, and anything you write into a consent flow is worth running past a lawyer once.

What this looks like in practice

Start by splitting your SMS reporting in two. One number for campaigns, one for flows, each with revenue per recipient alongside it. Most brands have never looked at the channel this way and the first view is usually uncomfortable, because the campaign half absorbs most of the send volume and most of the cost while the flow half quietly produces the revenue.

Then cut the promotional calendar to what you can justify and put the freed budget into the four triggers above. Hold cadence near the median of around two messages per subscriber per month while you build, and watch 30-day retention rather than revenue per send, because retention is the number that tells you whether you are building an asset or spending one.

Audit your consent trail last, before peak rather than after it. Check where every segment came from, confirm the opt-in form carries all five required elements, and delete any list you cannot trace to a one-to-one disclosure. Deleting subscribers feels like going backwards. It is cheaper than the alternative, and the numbers you lose were never converting anyway.

Inside the system

How we build this for brands

The flow build is the part that used to eat a month, so we automated it. Lifecycle sequences get drafted and deployed into Klaviyo by AI against the brand's own voice and its actual purchase data, which is what makes replenishment timing real rather than a 30-day guess: the reorder window comes per SKU out of live Shopify history instead of a default. The copy itself is drawn from our VOC engine, which mines reviews and support messages so the one line an SMS gets to use is phrased the way customers already describe the problem.

On top of that sits a reporting agent that keeps campaigns and flows split, tracks revenue per recipient and 30-day list retention weekly, and flags cadence drift before it shows up as churn. Part of this runs live for portfolio brands today; the full system is what we deploy when we take a brand on.

Frequently asked questions