Search for SMS marketing best practices and you will find the same advice repeated a dozen times. Get consent. Keep it short. Add a call to action. Send a coupon.
That advice is fine if you run an online store. It falls apart the moment you are texting a lead who requested an insurance quote, asked about a degree program, or started a loan application and went quiet. Those people aren’t looking for a coupon code. They want a reply, and they want it before your competitor gets to them.
Meera has been texting on behalf of B2C sales teams for years, and in 2025 alone we analyzed more than 35 million SMS interactions across insurance, financial services, and higher education to see what people reply to and when. We have also watched what happens when institutions like Penn Foster and lenders like Level Financing put these practices to work. This guide is a summation of what our team has learned.
Most best-practice lists treat compliance as one bullet near the bottom. That is backwards. A clever message that gets your number blocked by the carriers is worth nothing, and a TCPA claim can cost $500 to $1,500 per text. We have a full SMS compliance guide and a compliance checklist if you need the long version.
One more thing that is not a law but behaves like one. Apple and Google now filter messages from unknown senders into a separate folder. If you are not registered, not identifying yourself, and sending links in your first message, you may never reach the inbox at all. We covered what iOS 26 changed and how to adapt.
These run roughly in order of how much they move results for a sales team. The first few are table stakes. The last few are where the real gains hide.
Put a separate checkbox on your forms that says, in plain language, that the person agrees to receive text messages from your company, that message frequency varies, and that they can reply STOP at any time. Keep the timestamp and the form snapshot. If you buy leads, ask the vendor for that same proof for every record. Nothing else in this article matters if you cannot show consent.
Complete 10DLC registration before the first send. Then identify your business in every message, ideally in the first few words. People do not reply to mystery numbers. Pew Research found only 19% of Americans generally answer calls from numbers they do not recognize, and the same instinct applies to texts. "Hi Maria, this is Dan from Summit Insurance" beats "Hi Maria" every time.
Speed to lead is the single biggest lever most teams have. A lead who fills out a form is at peak interest for a few minutes. After that, interest decays and competitors show up. Level Financing built its outreach around this. New leads get a text within 15 seconds of opting in, and the results speak for themselves: 43% of leads reply, 56% of those become qualified, and 97% of qualified leads book a call. Read the full Level Financing story, or our guide to improving speed to lead.
This is the most important finding in our benchmark data. Across 35 million interactions, the opening message earned a 9.3% reply rate. The first follow-up earned 4.8%. The second earned 2.9%. The third earned 1.5%. Every follow-up is fishing in a smaller pond. If you only have time to optimize one thing, optimize the opener. Test the greeting, the question, and whether you reference the specific thing the person asked about.
Stay under 160 characters so the message arrives in one piece. Use the words a colleague would use. Ask one question or make one request, never three. Skip the text-speak, the all-caps, and the exclamation marks. And do not put a link in the first message. Links from unrecognized senders are a filtering trigger and they make you look like spam. Earn a reply first, then share the link.
Hi Jordan, this is Priya with Northgate Lending. Saw you started a personal loan application yesterday. Do you have 2 minutes to finish it, or would a quick call be easier?
Reply timestamps in our data cluster around noon in every industry we measured. Financial services showed a second spike between 3 and 4 p.m., which suggests those leads re-engage late in the workday. Early-morning sends, which a lot of outreach cadences default to, consistently underperformed. Your mileage will vary by audience, so test it. But if you are guessing, guess lunch.
Persistence works up to a point. On average it took 1.83 outbound attempts before a lead replied, which means most replies came on the first or second touch. After the second follow-up, response rates fall off a cliff and opt-outs climb. Two nudges after the opener is a sensible cap for most programs. If someone has not replied after three messages, park them and try again in a few weeks with a fresh angle.
A universal follow-up schedule does not exist. In our data, insurance leads replied in 3.2 days on average. Financial services took 5.9 days. Education took 6.4 days. So a next-day follow-up feels pushy to a prospective student who is still researching schools, and too slow for an insurance shopper who is comparing quotes this afternoon. The industry table further down gives you starting points.
One-way blasts train people to ignore you. A text that asks a question and gets a real reply within a minute trains them to engage. That means someone (or something) has to be watching the inbox. Set an internal standard for reply time, or automate the replies for the common questions so a human only steps in when it matters. Our guides to conversational texting and the problems with blasting go deeper.
Age and zip code tell you very little about what to say. Where the person is in your funnel tells you everything. A brand-new inquiry, a lead who went quiet after a quote, a no-show, and an applicant with an incomplete form each need a different opening line. Build your segments around those stages. Our guide to B2C lead management shows how to prioritize and route them.
Dropping in a first name is not personalization anymore. Referencing the program they asked about, the vehicle they quoted, or the appointment they missed is. That requires your texting platform to read from your CRM and write back to it, so the salesperson sees the whole conversation when the lead finally picks up the phone. If your tools do not integrate, you are guessing.
Penn Foster used to stop reaching out to any lead who had not responded within seven days. Those leads were converting at 1.5%. When the admissions team started texting them instead of calling, lead-to-enrollment climbed 42% and the school attributed $1.2 million in new revenue to leads it had previously abandoned. Life Chiropractic College West saw a 93.3% conversation rate on aged contacts. Old leads already know your name. That is a head start most new leads do not give you. See the Penn Foster and LifeWest case studies, then our aged lead strategies.
The point of the text conversation, for most sales teams, is a phone call. So design the handoff. When a lead says "yes, call me," the call should happen within minutes, ideally as a warm transfer while they are still holding the phone. Academy of Art University doubled its call transfer rate, from 20% to 40%, once texting handled the warm-up and admissions counselors only took calls from students who had already said yes. Our live call transfer guide explains how to set this up.
If 20% of your leads speak Spanish and 100% of your texts are in English, you are leaving replies on the table. Detect the language from the first reply and switch. It matters more than most teams expect, and it is easier than it used to be. We covered the mechanics in how to text in different languages.
You will see the claim everywhere that SMS has a 98% open rate. Trace that number back and it leads to a decade-old vendor statistic with no methodology behind it. Even if it were accurate, it would not tell you anything useful. Opens do not pay commissions. Track the metrics that map to revenue: reply rate on the first message, conversation rate (how many leads had a real back-and-forth), appointment or transfer rate, and opt-out rate. Opt-outs are your early warning system. A spike after a specific send means the copy, timing, or frequency was wrong.
Starting points, not rules. Test against your own data.
|
Industry |
First message |
Follow-up spacing |
Watch out for |
Go deeper |
|---|---|---|---|---|
|
Insurance |
Within 60 seconds of the quote request. Reference the coverage type. |
Day 1, day 3. Leads decide fast. |
State quiet hours (Florida 8 p.m. cutoff), DNC scrubbing on purchased leads |
|
|
Higher education |
Same day. Name the program and offer a specific next step. |
Day 2, day 7. Students research slowly. |
Sounding like a mass campaign; students spot it instantly |
|
|
Lending |
Within 15 seconds of opt-in. Ask if they want to finish the application. |
Day 1, day 4. Watch for month-end urgency. |
Anything that reads as a promise of approval |
|
|
Healthcare |
Confirmation on booking, reminder 48 hours out. |
One reminder 24 hours out, one no-show recovery. |
PHI in message content; keep it to logistics |
|
|
Home services |
Within minutes of the estimate request. Offer two time slots. |
Day 1, day 3. |
Late-evening sends after a long workday |
Each one identifies the sender, stays under 160 characters, asks one thing, and has no link in the opener. Swap the names and details. For more, see our 37 sales text examples.
Hi Alex, this is Sam from Ridgeline Insurance. Thanks for requesting an auto quote just now. Want me to text the numbers over, or is a quick call easier?
Hi Alex, Sam from Ridgeline Insurance. You looked at auto coverage with us a while back. Rates have shifted since then. Worth a fresh quote?
Hi Alex, Sam again from Ridgeline. No rush at all. If you still want that quote, just reply YES and I will send it over today.
Hi Alex, reminder that your call with Sam at Ridgeline is tomorrow at 2 p.m. Reply C to confirm or R to pick a new time.
Hi Alex, we missed you at 2 p.m. today. Totally understand, things come up. Would tomorrow at 10 a.m. or 3 p.m. work better?
You have been unsubscribed from Ridgeline Insurance texts and will not receive any more messages. Reply START to opt back in.
Look back at some of the best practices we listed above, such as:
* respond in seconds.
* Follow up on schedule and then stop.
* Answer every reply quickly.
* Hand off at the right moment.
Each of those is easy to do for ten leads and nearly impossible for a human team to do for ten thousand. That is the honest case for automation. Not that it writes better texts than your best rep, but that it does the tedious parts consistently, at 2 a.m., in the right time zone, without forgetting.
The caveat is that not all automation is equal. A drip tool that fires the same three messages at everyone is a blast with extra steps. What you want is a system that reads the reply, answers from content you have approved, knows the compliance rules for every state, and can tell the difference between "not interested" and "not right now."
Meera was built for exactly that. It sends the first message within seconds of a lead coming in, carries on the conversation from an approved knowledge base, respects quiet hours automatically, and transfers the lead to a live person the moment they say yes. You can see how it works in a few minutes.
Life Chiropractic College West is a good example of what that looks like in practice. Pandemic turnover left the admissions team with a single full-time staffer heading into its largest-ever winter class. Speed to contact was sitting above ten minutes. After deploying Meera, that lag dropped to near real time and the school recorded a 95.6% conversation rate, including 88.9% among students with previously canceled applications. One student showed up to a recruiting event and told the admissions director she already had an appointment booked. She had set it up over text. She did not realize she had been talking to software. Read the full LifeWest case study.
Midday. In our analysis of 35 million interactions, replies peaked around 12 p.m. across insurance, financial services, and education, with a secondary peak between 3 and 4 p.m. in financial services. Always send in the recipient's local time and stay inside 8 a.m. to 9 p.m., or tighter where state law requires it.
For marketing messages sent with any automation, no. The TCPA requires prior express written consent, and violations carry statutory damages of $500 to $1,500 per message. Transactional messages like appointment confirmations have more latitude, but consent is still the safe default.
160 characters for a single message using standard characters. Add an emoji or special character and the limit drops to 70. Longer messages are split into segments, which cost more and occasionally arrive out of order. Aim for 155 or fewer to leave room for a personalized name.
Nobody can say with confidence. The figure is repeated across thousands of vendor pages, but it traces back to old, unsourced industry claims rather than a published study. More importantly, open rate is not a useful metric for a sales team. Reply rate and booked calls are.
Two after the initial message. Our data shows replies arrive after 1.83 attempts on average, and response rates roughly halve with each additional follow-up. Beyond three total messages, you are mostly generating opt-outs.
If you take nothing else from this guide, take these. Get consent that names texting specifically and keep proof of it. Spend your creative energy on the first message, because it will earn more replies than every follow-up combined. And match your cadence to how your buyers actually behave, whether that is same-day for insurance or a patient week for education.
Everything else compounds on those three. If you want to see how your numbers compare, the full benchmark report is free to download. And if you would rather have someone else do the 2 a.m. follow-ups, book a demo and we will walk you through how Meera works.