SMS Marketing for D2C: Compliance Rules and Revenue Math That Actually Hold Up

Most brands turn on SMS because a case study said it prints money. Then the first legal letter shows up, or half the sends never reach a phone, and nobody on the team can say what a text actually earned.

SMS is not a copywriting channel. It is an operations channel with a legal layer on top, closer to running marketplace compliance than to writing an email subject line. The rules are federal. The filtering is done by phone carriers. The costs are per message, not per month. Get those three things wrong and a “high ROI channel” turns into a slow leak.

This guide covers what the law requires right now, how to find your real send limit, and how to run the math on what one subscriber is worth.

Quick Answer

What do D2C brands need to run SMS legally and profitably in the US?

• Get written opt-in for texts. It must be its own checkbox, separate from email.
• Register your number through A2P 10DLC before you send anything. Unregistered traffic gets filtered by carriers.
• Send only between 8 AM and 9 PM in the customer’s local time. Florida caps it at 8 PM.
• Honor opt-outs fast. The law gives you ten business days. Do it in minutes.
• Price every send at the true rate. Platform fee plus carrier fee runs about $0.010 to $0.020 per message segment.
• Find your send ceiling using your own churn math, not a blog rule of thumb.

Key takeaway: SMS usually earns more per subscriber than email. It also costs far more per send and carries real legal risk. Both facts are true at once, and only your own numbers settle it.

Why is SMS an ops problem and not a copy problem?

Email is cheap to send and nobody sues you for a newsletter. SMS is the opposite on both counts.

Three things make SMS different:

• Every send costs money. You pay per message segment. A long text splits into two segments and costs twice as much.
• Carriers can block you. AT&T, Verizon, and T-Mobile filter business texts. They do not owe you delivery.
• Bad consent creates legal exposure. Under federal law, a person who gets a text they never agreed to can sue you directly.

That mix means SMS needs an owner. Someone has to track consent records, watch delivery rates, manage the carrier registration, and clean the list. If nobody owns it, the program drifts until something breaks.

This is the same pattern we see across ecommerce operations that get outsourced too late. The channel works. The upkeep is what nobody staffed.

It is also why the dedicated team model beats the agency model for a channel like this. Agencies bill for campaigns. SMS breaks on the work between campaigns.

What are the SMS rules US brands must follow in 2026?

SMS compliance is not one rulebook. It is three, and they are enforced by different people for different reasons.

Layer	What it is	Who enforces it	What a breach costs you
TCPA	Federal law on consent and opt-out	FCC, plus private lawsuits	$500 to $1,500 per message
CTIA rules	Industry standards on content and format	The phone carriers	Filtering and number blocking
A2P 10DLC	Carrier registration for business texts	The Campaign Registry, carriers	Messages never arrive

The split matters. TCPA problems cost you money in court. CTIA problems cost you delivery. 10DLC problems mean your texts vanish. You can be clean on one and broken on another.

Here is where each stands as of August 2026.

You need written consent, and it must be separate

A phone number in your checkout field is not consent. The customer has to actively agree to receive marketing texts.

What counts:

• An unchecked box the customer ticks themselves
• Clear language saying they will get marketing texts
• A record of when, where, and how they agreed

What does not count:

• A pre-checked box
• A phone number given for shipping updates
• Their email opt-in carrying over to SMS

Keep the consent record. If you are ever challenged, the burden is on you to prove they said yes.

Quiet hours are 8 AM to 9 PM local

Federal rules limit marketing texts to the hours between 8 AM and 9 PM in the recipient’s time zone, per 47 CFR 64.1200.

Local time means their time, not yours. A 9 PM Eastern send hits California at 6 PM and Hawaii at 3 PM. That is fine. A 9 PM Pacific send hits the East Coast at midnight. That is a violation.

Several states are stricter than the federal floor. If you ship nationally, you are subject to all of them.

Where	Send window	What changes
Federal floor	8 AM to 9 PM local	Baseline
Florida	8 AM to 8 PM local	Cuts off an hour early
Oklahoma	8 AM to 8 PM local	Cuts off an hour early
Washington	8 AM to 8 PM local	Cuts off an hour early
Texas	9 AM to 9 PM Mon to Sat, 12 PM to 9 PM Sun	Later start, Sunday limits

Florida is the one that generates the most legal activity. The Florida Telephone Solicitation Act is the source. Texas added its own rules in September 2025 and may require registration with the Secretary of State before you send promotional texts to Texas residents.

Roughly fifteen states now layer extra rules on top of federal law. Trying to run a different rule per state is how mistakes happen. Set one policy at the tightest window you face and apply it everywhere.

Opt-outs must work, and you cannot force one method

Since April 11, 2025, you have to process an opt-out within ten business days. Any reasonable request counts.

The FCC named specific words that always count as an opt-out:

• stop
• quit
• end
• revoke
• opt out
• cancel
• unsubscribe

You also cannot tell people that one method is the only way out. If they reply “please stop texting me,” that works even though it is not on the list.

You may send one confirmation text. It has to go out within five minutes, carry no promotion, and make no attempt to talk them out of it.

Ten business days is the legal maximum. Treat it as an alarm, not a target. Your platform should handle it instantly.

What every message has to contain

These come from the carriers, not from federal law. Break them and your messages get filtered even if your consent is perfect.

• Your brand name. The person should never have to guess who is texting.
• A HELP keyword that works. If someone replies HELP, they must get your support contact back. This trips up more brands than any other item on this list.
• Opt-out wording. Include it in the first message of any sequence, and repeat it regularly after that.
• No public link shorteners. Bitly and TinyURL are strongly linked to spam traffic and trigger filters. Use a branded short domain.

The SHAFT rules on content

Carriers restrict five categories no matter what consent you hold. The shorthand is SHAFT.

• S is Sex. Explicit content is blocked.
• H is Hate. Blocked outright.
• A is Alcohol. Allowed with age gating.
• F is Firearms. Sales are prohibited.
• T is Tobacco. This also covers CBD and cannabis, with age gating required.

Related categories get flagged too, including gambling, payday loans, debt collection, and credit repair. If you sell in or near any of these, expect longer review during registration and ongoing filtering after it.

Two rule changes people keep getting wrong

Compliance content ages badly. Two items get repeated incorrectly across a lot of pages still ranking today.

The one-to-one consent rule is dead. The FCC wrote a rule that would have forced customers to consent to one company at a time. It was set to start January 27, 2025. On January 24, 2025, the Eleventh Circuit struck it down in Insurance Marketing Coalition v. FCC. The court said the FCC went past what the law allowed. Articles telling you to prepare for it are out of date.

The revoke-all rule has not started yet. This one would make a single opt-out stop every kind of message from you, even unrelated ones. It was supposed to begin April 11, 2025. The FCC pushed it to April 11, 2026, then pushed it again to January 31, 2027.

Do not treat that delay as a break. Build the cross-channel opt-out now. Retrofitting it under deadline pressure costs more than doing it right the first time.

The penalty math is what makes this serious

Damages run $500 per message, and $1,500 per message if the violation was willful. There is no cap on the total.

Individuals can sue you directly. You do not need to attract a regulator’s attention.

A single bad send to 5,000 people carries theoretical exposure of $2.5 million. That number is a ceiling, not a prediction. Real cases settle far below it.

What the real ones look like:

• Colony Ridge Development settled a texting-without-consent claim in 2025 for just under $2 million.
• Designer Brands settled for $4.4 million in 2025 over claims it kept texting people who had opted out.
• Athena Bitcoin settled for $4.5 million in 2026 over unsolicited promotional texts.

None of these needed huge send volume. They needed one gap in a consent or opt-out process.

Two more exposures most guides skip:

• The federal Do Not Call registry. Texting a registered number carries penalties far above the standard TCPA figure, into the tens of thousands per message.
• Reassigned numbers. Phone numbers get recycled. Texting a reassigned number means texting someone who never agreed. The FCC runs a Reassigned Numbers Database you can check against. Almost no D2C brand uses it. It is cheap insurance on an aging list.

This article is general information, not legal advice. Have a lawyer review your consent language and your opt-in flows before you launch.

What is A2P 10DLC and why do messages get filtered?

A2P 10DLC means “application to person, ten digit long code.” In plain terms, it is the system US carriers use to register business texting.

If you send automated texts from a normal ten digit number, you must register. You do this through The Campaign Registry, using your messaging platform as the go-between. You cannot register directly.

Registration means:

• Submitting your legal business details for verification
• Describing your campaign, with sample messages
• Paying a one-time brand fee plus a small monthly campaign fee
• Waiting. Approval commonly takes one to four weeks.

Skip it and carriers filter your traffic. Not “some messages go to spam.” Business texts do not have a spam folder. Filtered messages simply vanish. Your platform may still report them as sent.

Registration is also not a permanent pass. Carriers keep scoring your traffic after approval. Things that hurt your score:

• Sudden jumps in send volume
• High opt-out rates
• Shortened links that hide the real domain
• Heavy use of all caps and repeated punctuation
• Complaints from recipients

Diagnostic worth running monthly: compare messages sent to messages delivered, split by carrier. Treat filtered messages as wasted spend and fold them into your cost per order math. If one carrier’s delivery rate sits well below the others, that is a filtering problem, not a content problem. No amount of copy testing will fix it. You need to work the registration and reputation side.

How do you find your real send frequency ceiling?

Every guide says two to four texts per month. That number is guesswork applied to your business.

Your actual ceiling is the point where the people you lose outrun the people you add.

Here is the test.

Step one. Find your monthly opt-in rate. Count new SMS subscribers added last month.

Step two. Find your opt-out rate per send. Divide opt-outs by messages delivered, for each send.

Step three. Multiply. Opt-out rate times list size times number of sends equals monthly opt-outs.

Step four. Compare. If monthly opt-outs beat monthly opt-ins, your list is shrinking. You are past the ceiling.

Step five. Add one send per month and watch what happens to both numbers. Opt-out rate almost always climbs as frequency climbs. That is the signal to stop.

Worked example. A brand has 20,000 SMS subscribers and adds 900 new ones each month.

Sends per month	Opt-out rate per send	Monthly opt-outs	Net list change
4	0.40%	320	+580
6	0.50%	600	+300
8	0.65%	1,040	-140

At four sends the list grows. At eight it shrinks. The ceiling sits near six.

Note what changed. The opt-out rate itself rose with frequency. That is the part a fixed rule misses completely. You are not just sending more messages to the same list, you are making each message less welcome.

Run this every quarter. The ceiling moves as your list ages and your mix of subscribers changes.

This is one of the quiet limits on growth. Brands trying to scale from seven to eight figures often push send volume because it is the fastest lever they control. It works for a quarter, then the list stops replacing itself.

The THREAD system for running SMS as an operation

Six things, in order. Each one has an owner and a cadence.

T is for Track consent at the source. Log the timestamp, the page, and the exact wording shown. Store it where you can pull it in under a minute.

H is for Hit the register. Complete A2P 10DLC before your first send. Recheck your carrier delivery scores monthly.

R is for Ramp frequency slowly. Add one send per month at a time. Watch the opt-out rate. Stop when net list growth flattens.

E is for Earn the opt-in. Ask at moments where the customer already wants something from you. Checkout, order tracking, and back in stock alerts beat a homepage popup. A loyalty program gives you the strongest standing reason to ask.

A is for Attribute revenue net of cost. Every SMS report should subtract send cost. Gross revenue per send is a vanity number.

D is for Delete dead numbers. Remove subscribers who have not clicked or bought in six months. Also flush numbers you have not texted in 45 days or more, because disconnected numbers get reassigned to new people.

That last one matters more than it sounds. Texting a reassigned number means texting someone who never consented. That is exactly the situation the law was written for.

How much is one SMS subscriber actually worth?

Here is the full math, with real dollars. Same brand, 20,000 subscribers.

Step one: the true cost per send

Two costs stack, and most brands only count the first.

Cost line	Per SMS segment
Platform rate	$0.007 to $0.015
Carrier pass-through fee	$0.003 to $0.005
All-in per segment	$0.010 to $0.020

Carrier fees are set by AT&T, Verizon, and T-Mobile. No platform can waive them, and they went up again in January 2026. Add a small monthly campaign fee on top.

Use $0.014 per segment for this example.

Watch the segment count. A standard text is 160 characters. Go one character over and you pay for two segments. A 200 character message doubles your cost for no extra reach.

Step two: monthly send cost

• 6 sends per month to 20,000 subscribers = 120,000 segments
• 120,000 × $0.014 = $1,680
• Plus campaign fees, roughly $20
• Total: $1,700 per month

Step three: revenue per send

Build this from your own funnel, not a benchmark.

Revenue per send = click rate × purchase rate of clickers × average order value

For this brand:

• Click rate: 6%
• Purchase rate among clickers: 8%
• Average order value: $65
• 0.06 × 0.08 × $65 = $0.312 per send

Step four: net contribution

• 120,000 sends × $0.312 = $37,440 gross revenue
• At 40% contribution margin = $14,976
• Minus $1,700 in send cost = $13,276 net contribution
• Divided by 20,000 subscribers = $0.66 per subscriber per month

That is your number. Not open rate. Not ROI multiple. Net dollars per subscriber per month.

Once you have it, it feeds straight into your LTV to CAC ratio. A channel that raises retained value per customer changes what you can afford to pay to acquire one.

If you are not sure what your contribution margin is, start with how to calculate contribution margin for ecommerce before running this.

Does SMS really beat email per subscriber?

Usually yes per subscriber, and usually no in total. Both parts matter.

Run the same math on email for the same brand. Email has 20,000 contacts too, at a flat $700 per month.

• 8 sends per month = 160,000 emails
• Click rate 2%, purchase rate among clickers 6%, AOV $65
• Revenue per send = 0.02 × 0.06 × $65 = $0.078
• 160,000 × $0.078 = $12,480 gross
• At 40% margin = $4,992, minus $700 = $4,292 net
• Divided by 20,000 = $0.21 per subscriber per month

Side by side:

Metric	SMS	Email
Net per subscriber, monthly	$0.66	$0.21
Variable cost per send	$0.014	Near zero
Cost if you double sends	Doubles	Flat
Legal exposure per bad send	High	Low

SMS wins about three to one per subscriber here. But look at the third row.

Email cost does not move when you send more. SMS cost is fully variable. That changes how each channel behaves when you scale.

There is also a size gap. Most brands have five to ten times more email addresses than phone numbers, because email opt-in is easier to get. A channel earning a third as much per person across ten times the people still produces more total revenue.

The real read: SMS is your highest value per person channel. Email is your highest total revenue channel. Fund both, and stop asking which one wins.

The split usually falls out by job. Email carries the long explanation and the images. SMS carries the time sensitive nudge. That is why SMS earns its cost most clearly on abandoned cart recovery, where minutes matter, and least clearly on general promotions.

If your email side is not built out yet, the sequencing matters. Get your core ecommerce email marketing flows live first. Email flows are cheaper to test and teach you what offers work before you start paying per message to learn the same thing.

Which SMS stats should you not trust?

A few numbers get repeated everywhere and do not survive a close look.

“SMS has a 98% open rate.” There is no way to track SMS opens. Email uses a tracking pixel. SMS has nothing equivalent. Carriers report delivery, not opens. This figure is either a delivery rate wearing the wrong label, or an estimate nobody can verify. Use click rate and delivery rate instead. Both are real and both are measurable.

“SMS returns $X for every $1 spent.” These are almost always vendor internal analyses across their own paying customers. That group is self-selected and the method is rarely published. Treat them as marketing, not research.

Benchmark conversion tables. Numbers pulled from one platform’s customer base tell you about that platform’s customers. Useful as a rough sanity check. Not useful as a target.

Your own last 90 days beats any published benchmark. That is the whole point of running the math above.

How do you grow an SMS list without buying trouble?

List growth is where compliance and revenue meet. A sloppy opt-in creates both legal risk and a low quality subscriber.

Best places to ask, ranked by intent:

• Checkout. Highest intent moment you have. Use a separate SMS checkbox, never a shared one.
• Order tracking page. Customers actively want shipping updates here. Be clear that marketing texts are separate.
• Back in stock alerts. They asked for a text. That is real consent for that purpose. Get separate agreement for promotions.
• Post purchase. They already bought. Pair the ask with a reason, like early access. This is the same window that works for post purchase cross sell and for review requests.
• Loyalty signup. Points give a clear reason to hand over a number.
• Subscription signup. If you run recurring revenue, renewal and delivery texts are genuinely wanted. Get separate consent for promotions.
• Site popup. Works, but produces the weakest subscribers of the group.

Every opt-in point needs the same disclosure block:

• Who is sending
• What kind of messages
• Roughly how often
• That message and data rates may apply
• How to stop

Two things to avoid. Do not buy or rent phone lists. Consent does not transfer. Do not import numbers customers gave you for shipping or support and start marketing to them. That is one of the most common ways brands end up on the wrong side of this.

If retention is the goal behind your SMS push, the channel is one lever among several. Turning first time buyers into repeat customers depends on more than message frequency.

What does a working SMS program look like week to week?

Here is the operating cadence. This is the part that decides whether the channel keeps working after month three.

Every send:

• Confirm quiet hours across all time zones
• Check character count against the 160 segment break
• Confirm opt-out language is present

Weekly:

• Review delivery rate by carrier
• Review opt-out rate against the prior send
• Process any manual opt-out requests that came in through support

Monthly:

• Recalculate net contribution per send
• Compare monthly opt-ins to monthly opt-outs
• Review carrier trust score and registration status

Quarterly:

• Re-run the frequency ceiling test
• Remove subscribers with no engagement in six months
• Flush numbers untouched for 45 days or more
• Audit consent records on a random sample of 50 subscribers

Annually:

• Legal review of all opt-in language
• Review state level rules for every state you ship to

Write these into documented SOPs your team can actually follow. A checklist that lives in one person’s head is not a process.

Note how many of these are recurring checks rather than creative work. That is the real staffing question with SMS. The copy takes an hour. The compliance and delivery upkeep runs all month.

Where do most D2C SMS programs actually break?

Five failure patterns, in the order we see them.

• Nobody owns it. Marketing writes the copy. Nobody watches delivery rates or consent records. Problems surface only when revenue drops.
• Frequency creeps. A good quarter leads to more sends. The opt-out rate climbs quietly. Six months later the list is smaller than it was.
• Cost is never subtracted. Reports show gross revenue attributed to SMS. Send cost sits in a different spreadsheet, if anywhere.
• Consent records are thin. The opt-in works but nothing is logged. There is no way to prove consent later.
• Support and marketing are disconnected. A customer asks support to stop texting. Support has no way to action it. Marketing keeps sending.

That last one is worth sitting with. Under FCC rules you cannot require a specific opt-out method. If someone tells your support agent to stop, that counts. Your support team needs a path to act on it, and your support KPIs should reflect it. If support is handled by an outside team, the opt-out path has to be written into their scope or it will not exist.

Frequently asked questions

What to do next

Pick the gap that applies to you.

• No SMS program yet. Start with 10DLC registration and a single checkout opt-in. Nothing else until those are clean.
• Running SMS with no cost tracking. Run the four step revenue math above on your last 90 days. Most brands are surprised.
• Suspect frequency is too high. Run the ceiling test. Compare monthly opt-ins to monthly opt-outs.
• No clear owner. That is the real problem. Fix it before optimizing anything else.

SMS rewards brands that treat it as infrastructure and punishes brands that treat it as a broadcast tool. The difference is not talent. It is whether someone is watching the numbers every week.

Most scaling D2C teams do not have a spare person for that. AcquireX builds dedicated offshore teams that own the recurring work end to end, from consent tracking and carrier delivery monitoring to the weekly reporting that tells you whether the channel is actually paying for itself.

Owned channels also sit next to paid. If SMS and email are carrying retention, that changes what your performance marketing can afford to spend on acquisition.

If SMS is on your roadmap and nobody has bandwidth to run it properly, talk to us about what a dedicated team would cover.