How to Never Miss a WhatsApp Message Again (Even at 3 AM)
Your business closes at 6pm. Customers message at 9pm. By morning, they've gone elsewhere. Here's how to build an after-hours responder on WhatsApp yourself.
You close at 6 PM. At 9:14 PM, someone messages your WhatsApp: "Hey, do you have any openings tomorrow for a haircut?"
You don't see it until 8 AM the next day. You reply: "Hi! Yes, we have a 2 PM and a 4 PM available."
Silence. They already booked somewhere else. They found a place that responded at 9:14 PM.
This happens every single night to every business with a WhatsApp number. It's not a technology problem. It's a response time problem. And the math is brutal.
A study by Lead Connect found that 78% of customers buy from the business that responds first. Not the best business. Not the cheapest. The first one to reply.
When your competition responds at 9:14 PM and you respond at 8 AM, you've already lost. Not because you're worse. Because you were asleep.
The good news: closing that gap is now a build problem with a clear, short path. You don't buy a black box that "handles your DMs." You wire up a responder yourself, on infrastructure that gives you the primitives to do it in an afternoon.
Option 1: The WhatsApp Away Message (Free, but Useless)
WhatsApp Business has a built-in away message feature. It's free. It's easy to set up. And it's almost entirely useless.
Here's what it does: when someone messages you outside business hours, they get a single, pre-written message. Something like:
"Thanks for reaching out! We're currently closed. Our business hours are Mon-Sat 9 AM - 6 PM. We'll get back to you as soon as we can!"
That's it. One message. Same message for everyone. Whether they asked about pricing, availability, your location, or whether you can fit them in tomorrow.
What's wrong with this:
- It doesn't answer the question. The customer asked if you have openings tomorrow. They got back a message about your business hours. Those are two different things.
- It's a dead end. There's nothing for the customer to do except wait. No next step. No action. Just "we'll get back to you." Which really means "go find someone else."
- Everyone knows it's automated. And not in a good way. It feels like talking to a wall.
- You can't program it. There's no logic, no branching, no data lookup, no way to connect it to your calendar or your stack. It's a string, not a system.
The away message was designed for a world where customers were patient. That world doesn't exist anymore.
Option 2: A Responder You Build Yourself
Here's the same scenario, but this time you've built a real responder on top of the messaging API. It receives the inbound message, runs your logic, and replies, all in seconds:
Customer (9:14 PM): "Hey, do you have any openings tomorrow for a haircut?"
Your responder: "Hi! Let me check tomorrow's availability for you. We have:
- 10:30 AM with Sarah
- 1:00 PM with Mike
- 3:30 PM with Sarah
Would any of those work for you?"
Customer: "1 PM with Mike sounds great"
Your responder: "Done! You're booked:
Men's Haircut with Mike Tomorrow at 1:00 PM Duration: 30 minutes Price: $30
I'll send you a reminder in the morning. See you tomorrow!"
It's 9:16 PM. The customer is booked. You're watching a show. Your morning starts with a confirmed appointment instead of a missed opportunity.
This isn't hypothetical. It's exactly what your code does when you give it your services, your schedule, your pricing, and your staff availability. It doesn't guess. It checks your real data and gives a real answer, because you wrote it to.
The Difference in One Sentence
A WhatsApp away message says: "We're closed."
A responder you build says: "We open at 9 AM. Sarah has a 10:30 slot open tomorrow -- want me to book it?"
One is a wall. The other is a door. Customers walk through doors.
How You Build It on Wabery
Wabery is the messaging API you build on top of WhatsApp, Instagram, and Messenger. It gives you the primitives, a unified channels API, signed event webhooks, native WhatsApp Flows, automations, a CLI, and an MCP server, so you can build exactly the after-hours responder your business needs.
The shape of it is simple. Every inbound message hits your webhook with a verified signature. Your code decides what to send back:
// Receive every DM, reply with real answers
wabery.on("message.received", async (event) => {
const reply = await answerFor(event.text); // your logic + your data
await wabery.messages.send({
channel: event.channel, // whatsapp | instagram | messenger
to: event.from,
text: reply,
});
});
answerFor is yours. It can be a few if statements over keywords, a lookup against your calendar, or a call to an LLM with your services as context. Start small and grow it. The point is that you own the behavior, not a vendor.
If you'd rather not write the dispatcher at all, the CLI and built-in automations let you wire keyword and multi-step replies without code, and the MCP server lets an AI agent drive the whole conversation. You pick the level of control you want.
What This Looks Like After Hours: Real Examples
Let's walk through a few scenarios that happen every single night at businesses with WhatsApp numbers. In each, the away message is the dead end, and the responder you built is the alternative.
The Price Check at 10 PM
Customer: "How much do you charge for a full set of acrylic nails?"
Away message: "Thanks for reaching out! We'll get back to you during business hours."
Your responder: "A full set of acrylic nails is $55 and takes about 90 minutes. We also have gel extensions at $65 if you prefer a lighter feel. Would you like to book an appointment?"
The away message makes the customer wait until tomorrow. By then, they've googled three other nail salons and already know their prices. Your responder gives them the answer and moves them toward a booking in the same breath, because you wrote the pricing lookup into it.
The "Is This Even Possible?" Question at 11 PM
Customer: "I have a wedding next Saturday and need hair and makeup for me and two bridesmaids. Is that something you do?"
Away message: "Thanks for reaching out! We'll get back to you during business hours."
Your responder: "Congratulations! Yes, we do bridal hair and makeup packages. For a bride plus two bridesmaids, I'd recommend our Bridal Party Package:
- Bride: Hair styling + full makeup ($250)
- Each bridesmaid: Hair styling + makeup ($150)
Total for the three of you: $550. The session takes about 3-4 hours, so we'd typically start around 8 AM for a midday wedding.
Would you like me to check availability for next Saturday?"
The customer with a wedding isn't shopping casually. She's anxious, planning under pressure, and needs answers now. An away message tells her to wait. The responder you built gives her a plan and a path forward. That's a $550 booking that would have gone to whoever responded first.
The Reschedule at Midnight
Customer (12:17 AM): "I have an appointment tomorrow at 2 PM but something came up. Can I move it to Thursday?"
Away message: "Thanks for reaching out! We'll get back to you during business hours."
By morning, the customer might just no-show. They tried to do the right thing and reschedule, but they hit a wall.
Your responder: "No problem! Let me check Thursday for you. We have:
- 11:00 AM with Sarah
- 2:30 PM with Sarah
- 4:00 PM with Mike
Which time works best?"
The booking stays on your calendar instead of becoming a no-show, because your code reached into your calendar and offered the open slots. The customer feels taken care of instead of frustrated. And it happened at 12:17 AM.
Why Speed Matters More Than You Think
There's a psychological phenomenon at play here. When a customer messages a business, they're in a moment of intent. They want something right now. Not abstractly. Right now.
That intent has a half-life. Every minute that passes, the urgency fades. After 5 minutes, they're checking other options. After 30 minutes, they've moved on mentally. After 8 hours (a typical overnight delay), the moment is gone entirely.
The businesses that win aren't the ones with the best services or the lowest prices. They're the ones that are there when the customer is ready.
A responder you build is always there. At 9 PM on a Tuesday. At 6 AM on a Sunday. At midnight on New Year's Eve. Not "there" in the away-message sense of acknowledging existence. Actually there -- answering questions, checking availability, making bookings, because that's the code you shipped.
Setting This Up Takes an Afternoon
This isn't a 3-week implementation project. Here's the actual path:
- Create a Wabery account. Free. No credit card.
- Connect your WhatsApp number. QR code scan, same as WhatsApp Web.
- Point a webhook at your endpoint (or use the CLI to set up no-code rules). Wabery verifies the signature on every event.
- Write your reply logic. Start with keyword rules for the common after-hours questions: pricing, hours, services, basic availability. Grow it from there.
The free plan is enough to build and test the whole loop. When you want richer, multi-turn conversations, the MCP server lets you hand the conversation to an AI agent that reasons over your data, and you stay in control of when it does.
But start by building the simple version tonight. See what happens when your 9 PM messages get real answers instead of away messages.
What About the Messages You Should Handle Personally?
Not every message should get an automated response. Some conversations need your personal touch -- a loyal customer with a specific request, a complaint, a question about a specialized service.
Because you wrote the logic, you decide the boundary. Detect the cases that are outside scope, reply appropriately, and route them to your team. You see every conversation and can jump in at any time. The webhook gives you the full event stream, so building a "hand off to a human" path is just another branch in your code.
The system handles the routine. You handle the exceptions. That's the right division of labor, and you draw the line exactly where it makes sense for your business.
The Overnight Test
Try this tonight. Set up Wabery on the free plan. It takes an afternoon. Connect your WhatsApp. Wire a webhook and a few reply rules.
Then go to sleep.
Check your dashboard in the morning. Look at the messages that came in overnight. Look at the responses your code sent. Look at whether customers got answers or got away messages.
If you want to go deeper on how a responder you build differs from a traditional WhatsApp chatbot, read AI Agent vs Chatbot: What's the Difference. And if the free plan convinces you, here's what you can build with the full toolkit.
One night. That's all it takes to see the difference.
Questions or feedback? Reach out anytime